diff options
Diffstat (limited to 'utils/text')
| -rw-r--r-- | utils/text/Kyuafile | 9 | ||||
| -rw-r--r-- | utils/text/Makefile.am.inc | 74 | ||||
| -rw-r--r-- | utils/text/exceptions.cpp | 91 | ||||
| -rw-r--r-- | utils/text/exceptions.hpp | 77 | ||||
| -rw-r--r-- | utils/text/exceptions_test.cpp | 76 | ||||
| -rw-r--r-- | utils/text/operations.cpp | 261 | ||||
| -rw-r--r-- | utils/text/operations.hpp | 68 | ||||
| -rw-r--r-- | utils/text/operations.ipp | 91 | ||||
| -rw-r--r-- | utils/text/operations_test.cpp | 435 | ||||
| -rw-r--r-- | utils/text/regex.cpp | 302 | ||||
| -rw-r--r-- | utils/text/regex.hpp | 92 | ||||
| -rw-r--r-- | utils/text/regex_fwd.hpp | 46 | ||||
| -rw-r--r-- | utils/text/regex_test.cpp | 177 | ||||
| -rw-r--r-- | utils/text/table.cpp | 428 | ||||
| -rw-r--r-- | utils/text/table.hpp | 125 | ||||
| -rw-r--r-- | utils/text/table_fwd.hpp | 58 | ||||
| -rw-r--r-- | utils/text/table_test.cpp | 413 | ||||
| -rw-r--r-- | utils/text/templates.cpp | 764 | ||||
| -rw-r--r-- | utils/text/templates.hpp | 122 | ||||
| -rw-r--r-- | utils/text/templates_fwd.hpp | 45 | ||||
| -rw-r--r-- | utils/text/templates_test.cpp | 1001 |
21 files changed, 4755 insertions, 0 deletions
diff --git a/utils/text/Kyuafile b/utils/text/Kyuafile new file mode 100644 index 000000000000..e4e870e9c648 --- /dev/null +++ b/utils/text/Kyuafile @@ -0,0 +1,9 @@ +syntax(2) + +test_suite("kyua") + +atf_test_program{name="exceptions_test"} +atf_test_program{name="operations_test"} +atf_test_program{name="regex_test"} +atf_test_program{name="table_test"} +atf_test_program{name="templates_test"} diff --git a/utils/text/Makefile.am.inc b/utils/text/Makefile.am.inc new file mode 100644 index 000000000000..d474ae191bf5 --- /dev/null +++ b/utils/text/Makefile.am.inc @@ -0,0 +1,74 @@ +# Copyright 2012 The Kyua Authors. +# All rights reserved. +# +# Redistribution and use in source and binary forms, with or without +# modification, are permitted provided that the following conditions are +# met: +# +# * Redistributions of source code must retain the above copyright +# notice, this list of conditions and the following disclaimer. +# * Redistributions in binary form must reproduce the above copyright +# notice, this list of conditions and the following disclaimer in the +# documentation and/or other materials provided with the distribution. +# * Neither the name of Google Inc. nor the names of its contributors +# may be used to endorse or promote products derived from this software +# without specific prior written permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +libutils_a_SOURCES += utils/text/exceptions.cpp +libutils_a_SOURCES += utils/text/exceptions.hpp +libutils_a_SOURCES += utils/text/operations.cpp +libutils_a_SOURCES += utils/text/operations.hpp +libutils_a_SOURCES += utils/text/operations.ipp +libutils_a_SOURCES += utils/text/regex.cpp +libutils_a_SOURCES += utils/text/regex.hpp +libutils_a_SOURCES += utils/text/regex_fwd.hpp +libutils_a_SOURCES += utils/text/table.cpp +libutils_a_SOURCES += utils/text/table.hpp +libutils_a_SOURCES += utils/text/table_fwd.hpp +libutils_a_SOURCES += utils/text/templates.cpp +libutils_a_SOURCES += utils/text/templates.hpp +libutils_a_SOURCES += utils/text/templates_fwd.hpp + +if WITH_ATF +tests_utils_textdir = $(pkgtestsdir)/utils/text + +tests_utils_text_DATA = utils/text/Kyuafile +EXTRA_DIST += $(tests_utils_text_DATA) + +tests_utils_text_PROGRAMS = utils/text/exceptions_test +utils_text_exceptions_test_SOURCES = utils/text/exceptions_test.cpp +utils_text_exceptions_test_CXXFLAGS = $(UTILS_CFLAGS) $(ATF_CXX_CFLAGS) +utils_text_exceptions_test_LDADD = $(UTILS_LIBS) $(ATF_CXX_LIBS) + +tests_utils_text_PROGRAMS += utils/text/operations_test +utils_text_operations_test_SOURCES = utils/text/operations_test.cpp +utils_text_operations_test_CXXFLAGS = $(UTILS_CFLAGS) $(ATF_CXX_CFLAGS) +utils_text_operations_test_LDADD = $(UTILS_LIBS) $(ATF_CXX_LIBS) + +tests_utils_text_PROGRAMS += utils/text/regex_test +utils_text_regex_test_SOURCES = utils/text/regex_test.cpp +utils_text_regex_test_CXXFLAGS = $(UTILS_CFLAGS) $(ATF_CXX_CFLAGS) +utils_text_regex_test_LDADD = $(UTILS_LIBS) $(ATF_CXX_LIBS) + +tests_utils_text_PROGRAMS += utils/text/table_test +utils_text_table_test_SOURCES = utils/text/table_test.cpp +utils_text_table_test_CXXFLAGS = $(UTILS_CFLAGS) $(ATF_CXX_CFLAGS) +utils_text_table_test_LDADD = $(UTILS_LIBS) $(ATF_CXX_LIBS) + +tests_utils_text_PROGRAMS += utils/text/templates_test +utils_text_templates_test_SOURCES = utils/text/templates_test.cpp +utils_text_templates_test_CXXFLAGS = $(UTILS_CFLAGS) $(ATF_CXX_CFLAGS) +utils_text_templates_test_LDADD = $(UTILS_LIBS) $(ATF_CXX_LIBS) +endif diff --git a/utils/text/exceptions.cpp b/utils/text/exceptions.cpp new file mode 100644 index 000000000000..1692cfea7edb --- /dev/null +++ b/utils/text/exceptions.cpp @@ -0,0 +1,91 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/exceptions.hpp" + +namespace text = utils::text; + + +/// Constructs a new error with a plain-text message. +/// +/// \param message The plain-text error message. +text::error::error(const std::string& message) : + std::runtime_error(message) +{ +} + + +/// Destructor for the error. +text::error::~error(void) throw() +{ +} + + +/// Constructs a new error with a plain-text message. +/// +/// \param message The plain-text error message. +text::regex_error::regex_error(const std::string& message) : + error(message) +{ +} + + +/// Destructor for the error. +text::regex_error::~regex_error(void) throw() +{ +} + + +/// Constructs a new error with a plain-text message. +/// +/// \param message The plain-text error message. +text::syntax_error::syntax_error(const std::string& message) : + error(message) +{ +} + + +/// Destructor for the error. +text::syntax_error::~syntax_error(void) throw() +{ +} + + +/// Constructs a new error with a plain-text message. +/// +/// \param message The plain-text error message. +text::value_error::value_error(const std::string& message) : + error(message) +{ +} + + +/// Destructor for the error. +text::value_error::~value_error(void) throw() +{ +} diff --git a/utils/text/exceptions.hpp b/utils/text/exceptions.hpp new file mode 100644 index 000000000000..da0cfd98fb88 --- /dev/null +++ b/utils/text/exceptions.hpp @@ -0,0 +1,77 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/exceptions.hpp +/// Exception types raised by the text module. + +#if !defined(UTILS_TEXT_EXCEPTIONS_HPP) +#define UTILS_TEXT_EXCEPTIONS_HPP + +#include <stdexcept> + +namespace utils { +namespace text { + + +/// Base exceptions for text errors. +class error : public std::runtime_error { +public: + explicit error(const std::string&); + ~error(void) throw(); +}; + + +/// Exception denoting an error in a regular expression. +class regex_error : public error { +public: + explicit regex_error(const std::string&); + ~regex_error(void) throw(); +}; + + +/// Exception denoting an error while parsing templates. +class syntax_error : public error { +public: + explicit syntax_error(const std::string&); + ~syntax_error(void) throw(); +}; + + +/// Exception denoting an error in a text value format. +class value_error : public error { +public: + explicit value_error(const std::string&); + ~value_error(void) throw(); +}; + + +} // namespace text +} // namespace utils + + +#endif // !defined(UTILS_TEXT_EXCEPTIONS_HPP) diff --git a/utils/text/exceptions_test.cpp b/utils/text/exceptions_test.cpp new file mode 100644 index 000000000000..1d3c3910900a --- /dev/null +++ b/utils/text/exceptions_test.cpp @@ -0,0 +1,76 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/exceptions.hpp" + +#include <cstring> + +#include <atf-c++.hpp> + +namespace text = utils::text; + + +ATF_TEST_CASE_WITHOUT_HEAD(error); +ATF_TEST_CASE_BODY(error) +{ + const text::error e("Some text"); + ATF_REQUIRE(std::strcmp("Some text", e.what()) == 0); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(regex_error); +ATF_TEST_CASE_BODY(regex_error) +{ + const text::regex_error e("Some text"); + ATF_REQUIRE(std::strcmp("Some text", e.what()) == 0); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(syntax_error); +ATF_TEST_CASE_BODY(syntax_error) +{ + const text::syntax_error e("Some text"); + ATF_REQUIRE(std::strcmp("Some text", e.what()) == 0); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(value_error); +ATF_TEST_CASE_BODY(value_error) +{ + const text::value_error e("Some text"); + ATF_REQUIRE(std::strcmp("Some text", e.what()) == 0); +} + + +ATF_INIT_TEST_CASES(tcs) +{ + ATF_ADD_TEST_CASE(tcs, error); + ATF_ADD_TEST_CASE(tcs, regex_error); + ATF_ADD_TEST_CASE(tcs, syntax_error); + ATF_ADD_TEST_CASE(tcs, value_error); +} diff --git a/utils/text/operations.cpp b/utils/text/operations.cpp new file mode 100644 index 000000000000..5a4345d979c7 --- /dev/null +++ b/utils/text/operations.cpp @@ -0,0 +1,261 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/operations.ipp" + +#include <sstream> + +#include "utils/format/macros.hpp" +#include "utils/sanity.hpp" + +namespace text = utils::text; + + +/// Replaces XML special characters from an input string. +/// +/// The list of XML special characters is specified here: +/// http://www.w3.org/TR/xml11/#charsets +/// +/// \param in The input to quote. +/// +/// \return A quoted string without any XML special characters. +std::string +text::escape_xml(const std::string& in) +{ + std::ostringstream quoted; + + for (std::string::const_iterator it = in.begin(); + it != in.end(); ++it) { + unsigned char c = (unsigned char)*it; + if (c == '"') { + quoted << """; + } else if (c == '&') { + quoted << "&"; + } else if (c == '<') { + quoted << "<"; + } else if (c == '>') { + quoted << ">"; + } else if (c == '\'') { + quoted << "'"; + } else if ((c >= 0x01 && c <= 0x08) || + (c >= 0x0B && c <= 0x0C) || + (c >= 0x0E && c <= 0x1F) || + (c >= 0x7F && c <= 0x84) || + (c >= 0x86 && c <= 0x9F)) { + // for RestrictedChar characters, escape them + // as '&#[decimal ASCII value];' + // so that in the XML file we will see the escaped + // character. + quoted << "&#" << static_cast< std::string::size_type >(*it) + << ";"; + } else { + quoted << *it; + } + } + return quoted.str(); +} + + +/// Surrounds a string with quotes, escaping the quote itself if needed. +/// +/// \param text The string to quote. +/// \param quote The quote character to use. +/// +/// \return The quoted string. +std::string +text::quote(const std::string& text, const char quote) +{ + std::ostringstream quoted; + quoted << quote; + + std::string::size_type start_pos = 0; + std::string::size_type last_pos = text.find(quote); + while (last_pos != std::string::npos) { + quoted << text.substr(start_pos, last_pos - start_pos) << '\\'; + start_pos = last_pos; + last_pos = text.find(quote, start_pos + 1); + } + quoted << text.substr(start_pos); + + quoted << quote; + return quoted.str(); +} + + +/// Fills a paragraph to the specified length. +/// +/// This preserves any sequence of spaces in the input and any possible +/// newlines. Sequences of spaces may be split in half (and thus one space is +/// lost), but the rest of the spaces will be preserved as either trailing or +/// leading spaces. +/// +/// \param input The string to refill. +/// \param target_width The width to refill the paragraph to. +/// +/// \return The refilled paragraph as a sequence of independent lines. +std::vector< std::string > +text::refill(const std::string& input, const std::size_t target_width) +{ + std::vector< std::string > output; + + std::string::size_type start = 0; + while (start < input.length()) { + std::string::size_type width; + if (start + target_width >= input.length()) + width = input.length() - start; + else { + if (input[start + target_width] == ' ') { + width = target_width; + } else { + const std::string::size_type pos = input.find_last_of( + " ", start + target_width - 1); + if (pos == std::string::npos || pos < start + 1) { + width = input.find_first_of(" ", start + target_width); + if (width == std::string::npos) + width = input.length() - start; + else + width -= start; + } else { + width = pos - start; + } + } + } + INV(width != std::string::npos); + INV(start + width <= input.length()); + INV(input[start + width] == ' ' || input[start + width] == '\0'); + output.push_back(input.substr(start, width)); + + start += width + 1; + } + + if (input.empty()) { + INV(output.empty()); + output.push_back(""); + } + + return output; +} + + +/// Fills a paragraph to the specified length. +/// +/// See the documentation for refill() for additional details. +/// +/// \param input The string to refill. +/// \param target_width The width to refill the paragraph to. +/// +/// \return The refilled paragraph as a string with embedded newlines. +std::string +text::refill_as_string(const std::string& input, const std::size_t target_width) +{ + return join(refill(input, target_width), "\n"); +} + + +/// Replaces all occurrences of a substring in a string. +/// +/// \param input The string in which to perform the replacement. +/// \param search The pattern to be replaced. +/// \param replacement The substring to replace search with. +/// +/// \return A copy of input with the replacements performed. +std::string +text::replace_all(const std::string& input, const std::string& search, + const std::string& replacement) +{ + std::string output; + + std::string::size_type pos, lastpos = 0; + while ((pos = input.find(search, lastpos)) != std::string::npos) { + output += input.substr(lastpos, pos - lastpos); + output += replacement; + lastpos = pos + search.length(); + } + output += input.substr(lastpos); + + return output; +} + + +/// Splits a string into different components. +/// +/// \param str The string to split. +/// \param delimiter The separator to use to split the words. +/// +/// \return The different words in the input string as split by the provided +/// delimiter. +std::vector< std::string > +text::split(const std::string& str, const char delimiter) +{ + std::vector< std::string > words; + if (!str.empty()) { + std::string::size_type pos = str.find(delimiter); + words.push_back(str.substr(0, pos)); + while (pos != std::string::npos) { + ++pos; + const std::string::size_type next = str.find(delimiter, pos); + words.push_back(str.substr(pos, next - pos)); + pos = next; + } + } + return words; +} + + +/// Converts a string to a boolean. +/// +/// \param str The string to convert. +/// +/// \return The converted string, if the input string was valid. +/// +/// \throw std::value_error If the input string does not represent a valid +/// boolean value. +template<> +bool +text::to_type(const std::string& str) +{ + if (str == "true") + return true; + else if (str == "false") + return false; + else + throw value_error(F("Invalid boolean value '%s'") % str); +} + + +/// Identity function for to_type, for genericity purposes. +/// +/// \param str The string to convert. +/// +/// \return The input string. +template<> +std::string +text::to_type(const std::string& str) +{ + return str; +} diff --git a/utils/text/operations.hpp b/utils/text/operations.hpp new file mode 100644 index 000000000000..6d15be553b06 --- /dev/null +++ b/utils/text/operations.hpp @@ -0,0 +1,68 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/operations.hpp +/// Utilities to manipulate strings. + +#if !defined(UTILS_TEXT_OPERATIONS_HPP) +#define UTILS_TEXT_OPERATIONS_HPP + +#include <cstddef> +#include <string> +#include <vector> + +namespace utils { +namespace text { + + +std::string escape_xml(const std::string&); +std::string quote(const std::string&, const char); + + +std::vector< std::string > refill(const std::string&, const std::size_t); +std::string refill_as_string(const std::string&, const std::size_t); + +std::string replace_all(const std::string&, const std::string&, + const std::string&); + +template< typename Collection > +std::string join(const Collection&, const std::string&); +std::vector< std::string > split(const std::string&, const char); + +template< typename Type > +Type to_type(const std::string&); +template<> +bool to_type(const std::string&); +template<> +std::string to_type(const std::string&); + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_OPERATIONS_HPP) diff --git a/utils/text/operations.ipp b/utils/text/operations.ipp new file mode 100644 index 000000000000..511cd6840a08 --- /dev/null +++ b/utils/text/operations.ipp @@ -0,0 +1,91 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#if !defined(UTILS_TEXT_OPERATIONS_IPP) +#define UTILS_TEXT_OPERATIONS_IPP + +#include "utils/text/operations.hpp" + +#include <sstream> + +#include "utils/text/exceptions.hpp" + + +/// Concatenates a collection of strings into a single string. +/// +/// \param strings The collection of strings to concatenate. If the collection +/// is unordered, the ordering in the output is undefined. +/// \param delimiter The delimiter to use to separate the strings. +/// +/// \return The concatenated strings. +template< typename Collection > +std::string +utils::text::join(const Collection& strings, const std::string& delimiter) +{ + std::ostringstream output; + if (strings.size() > 1) { + for (typename Collection::const_iterator iter = strings.begin(); + iter != --strings.end(); ++iter) + output << (*iter) << delimiter; + } + if (strings.size() > 0) + output << *(--strings.end()); + return output.str(); +} + + +/// Converts a string to a native type. +/// +/// \tparam Type The type to convert the string to. An input stream operator +/// must exist to extract such a type from an std::istream. +/// \param str The string to convert. +/// +/// \return The converted string, if the input string was valid. +/// +/// \throw std::value_error If the input string does not represent a valid +/// target type. This exception does not include any details, so the caller +/// must take care to re-raise it with appropriate details. +template< typename Type > +Type +utils::text::to_type(const std::string& str) +{ + if (str.empty()) + throw text::value_error("Empty string"); + if (str[0] == ' ') + throw text::value_error("Invalid value"); + + std::istringstream input(str); + Type value; + input >> value; + if (!input.eof() || input.bad() || input.fail()) + throw text::value_error("Invalid value"); + return value; +} + + +#endif // !defined(UTILS_TEXT_OPERATIONS_IPP) diff --git a/utils/text/operations_test.cpp b/utils/text/operations_test.cpp new file mode 100644 index 000000000000..2d5ab36c9090 --- /dev/null +++ b/utils/text/operations_test.cpp @@ -0,0 +1,435 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/operations.ipp" + +#include <iostream> +#include <set> +#include <string> +#include <vector> + +#include <atf-c++.hpp> + +#include "utils/text/exceptions.hpp" + +namespace text = utils::text; + + +namespace { + + +/// Tests text::refill() on an input string with a range of widths. +/// +/// \param expected The expected refilled paragraph. +/// \param input The input paragraph to be refilled. +/// \param first_width The first width to validate. +/// \param last_width The last width to validate (inclusive). +static void +refill_test(const char* expected, const char* input, + const std::size_t first_width, const std::size_t last_width) +{ + for (std::size_t width = first_width; width <= last_width; ++width) { + const std::vector< std::string > lines = text::split(expected, '\n'); + std::cout << "Breaking at width " << width << '\n'; + ATF_REQUIRE_EQ(expected, text::refill_as_string(input, width)); + ATF_REQUIRE(lines == text::refill(input, width)); + } +} + + +} // anonymous namespace + + +ATF_TEST_CASE_WITHOUT_HEAD(escape_xml__empty); +ATF_TEST_CASE_BODY(escape_xml__empty) +{ + ATF_REQUIRE_EQ("", text::escape_xml("")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(escape_xml__no_escaping); +ATF_TEST_CASE_BODY(escape_xml__no_escaping) +{ + ATF_REQUIRE_EQ("a", text::escape_xml("a")); + ATF_REQUIRE_EQ("Some text!", text::escape_xml("Some text!")); + ATF_REQUIRE_EQ("\n\t\r", text::escape_xml("\n\t\r")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(escape_xml__some_escaping); +ATF_TEST_CASE_BODY(escape_xml__some_escaping) +{ + ATF_REQUIRE_EQ("'", text::escape_xml("'")); + + ATF_REQUIRE_EQ("foo "bar& <tag> yay' baz", + text::escape_xml("foo \"bar& <tag> yay' baz")); + + ATF_REQUIRE_EQ(""&<>'", text::escape_xml("\"&<>'")); + ATF_REQUIRE_EQ("&&&", text::escape_xml("&&&")); + ATF_REQUIRE_EQ("&#8;&#11;", text::escape_xml("\b\v")); + ATF_REQUIRE_EQ("\t&#127;BAR&", text::escape_xml("\t\x7f""BAR&")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(quote__empty); +ATF_TEST_CASE_BODY(quote__empty) +{ + ATF_REQUIRE_EQ("''", text::quote("", '\'')); + ATF_REQUIRE_EQ("##", text::quote("", '#')); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(quote__no_escaping); +ATF_TEST_CASE_BODY(quote__no_escaping) +{ + ATF_REQUIRE_EQ("'Some text\"'", text::quote("Some text\"", '\'')); + ATF_REQUIRE_EQ("#Another'string#", text::quote("Another'string", '#')); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(quote__some_escaping); +ATF_TEST_CASE_BODY(quote__some_escaping) +{ + ATF_REQUIRE_EQ("'Some\\'text'", text::quote("Some'text", '\'')); + ATF_REQUIRE_EQ("#Some\\#text#", text::quote("Some#text", '#')); + + ATF_REQUIRE_EQ("'More than one\\' quote\\''", + text::quote("More than one' quote'", '\'')); + ATF_REQUIRE_EQ("'Multiple quotes \\'\\'\\' together'", + text::quote("Multiple quotes ''' together", '\'')); + + ATF_REQUIRE_EQ("'\\'escape at the beginning'", + text::quote("'escape at the beginning", '\'')); + ATF_REQUIRE_EQ("'escape at the end\\''", + text::quote("escape at the end'", '\'')); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__empty); +ATF_TEST_CASE_BODY(refill__empty) +{ + ATF_REQUIRE_EQ(1, text::refill("", 0).size()); + ATF_REQUIRE(text::refill("", 0)[0].empty()); + ATF_REQUIRE_EQ("", text::refill_as_string("", 0)); + + ATF_REQUIRE_EQ(1, text::refill("", 10).size()); + ATF_REQUIRE(text::refill("", 10)[0].empty()); + ATF_REQUIRE_EQ("", text::refill_as_string("", 10)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__no_changes); +ATF_TEST_CASE_BODY(refill__no_changes) +{ + std::vector< std::string > exp_lines; + exp_lines.push_back("foo bar\nbaz"); + + ATF_REQUIRE(exp_lines == text::refill("foo bar\nbaz", 12)); + ATF_REQUIRE_EQ("foo bar\nbaz", text::refill_as_string("foo bar\nbaz", 12)); + + ATF_REQUIRE(exp_lines == text::refill("foo bar\nbaz", 18)); + ATF_REQUIRE_EQ("foo bar\nbaz", text::refill_as_string("foo bar\nbaz", 80)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__break_one); +ATF_TEST_CASE_BODY(refill__break_one) +{ + refill_test("only break the\nfirst line", "only break the first line", + 14, 19); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__break_one__not_first_word); +ATF_TEST_CASE_BODY(refill__break_one__not_first_word) +{ + refill_test("first-long-word\nother\nwords", "first-long-word other words", + 6, 10); + refill_test("first-long-word\nother words", "first-long-word other words", + 11, 20); + refill_test("first-long-word other\nwords", "first-long-word other words", + 21, 26); + refill_test("first-long-word other words", "first-long-word other words", + 27, 28); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__break_many); +ATF_TEST_CASE_BODY(refill__break_many) +{ + refill_test("this is a long\nparagraph to be\nsplit into\npieces", + "this is a long paragraph to be split into pieces", + 15, 15); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__cannot_break); +ATF_TEST_CASE_BODY(refill__cannot_break) +{ + refill_test("this-is-a-long-string", "this-is-a-long-string", 5, 5); + + refill_test("this is\na-string-with-long-words", + "this is a-string-with-long-words", 10, 10); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(refill__preserve_whitespace); +ATF_TEST_CASE_BODY(refill__preserve_whitespace) +{ + refill_test("foo bar baz ", "foo bar baz ", 80, 80); + refill_test("foo \n bar", "foo bar", 5, 5); + + std::vector< std::string > exp_lines; + exp_lines.push_back("foo \n"); + exp_lines.push_back(" bar"); + ATF_REQUIRE(exp_lines == text::refill("foo \n bar", 5)); + ATF_REQUIRE_EQ("foo \n\n bar", text::refill_as_string("foo \n bar", 5)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(join__empty); +ATF_TEST_CASE_BODY(join__empty) +{ + std::vector< std::string > lines; + ATF_REQUIRE_EQ("", text::join(lines, " ")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(join__one); +ATF_TEST_CASE_BODY(join__one) +{ + std::vector< std::string > lines; + lines.push_back("first line"); + ATF_REQUIRE_EQ("first line", text::join(lines, "*")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(join__several); +ATF_TEST_CASE_BODY(join__several) +{ + std::vector< std::string > lines; + lines.push_back("first abc"); + lines.push_back("second"); + lines.push_back("and last line"); + ATF_REQUIRE_EQ("first abc second and last line", text::join(lines, " ")); + ATF_REQUIRE_EQ("first abc***second***and last line", + text::join(lines, "***")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(join__unordered); +ATF_TEST_CASE_BODY(join__unordered) +{ + std::set< std::string > lines; + lines.insert("first"); + lines.insert("second"); + const std::string joined = text::join(lines, " "); + ATF_REQUIRE(joined == "first second" || joined == "second first"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(split__empty); +ATF_TEST_CASE_BODY(split__empty) +{ + std::vector< std::string > words = text::split("", ' '); + std::vector< std::string > exp_words; + ATF_REQUIRE(exp_words == words); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(split__one); +ATF_TEST_CASE_BODY(split__one) +{ + std::vector< std::string > words = text::split("foo", ' '); + std::vector< std::string > exp_words; + exp_words.push_back("foo"); + ATF_REQUIRE(exp_words == words); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(split__several__simple); +ATF_TEST_CASE_BODY(split__several__simple) +{ + std::vector< std::string > words = text::split("foo bar baz", ' '); + std::vector< std::string > exp_words; + exp_words.push_back("foo"); + exp_words.push_back("bar"); + exp_words.push_back("baz"); + ATF_REQUIRE(exp_words == words); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(split__several__delimiters); +ATF_TEST_CASE_BODY(split__several__delimiters) +{ + std::vector< std::string > words = text::split("XfooXXbarXXXbazXX", 'X'); + std::vector< std::string > exp_words; + exp_words.push_back(""); + exp_words.push_back("foo"); + exp_words.push_back(""); + exp_words.push_back("bar"); + exp_words.push_back(""); + exp_words.push_back(""); + exp_words.push_back("baz"); + exp_words.push_back(""); + exp_words.push_back(""); + ATF_REQUIRE(exp_words == words); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(replace_all__empty); +ATF_TEST_CASE_BODY(replace_all__empty) +{ + ATF_REQUIRE_EQ("", text::replace_all("", "search", "replacement")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(replace_all__none); +ATF_TEST_CASE_BODY(replace_all__none) +{ + ATF_REQUIRE_EQ("string without matches", + text::replace_all("string without matches", + "WITHOUT", "replacement")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(replace_all__one); +ATF_TEST_CASE_BODY(replace_all__one) +{ + ATF_REQUIRE_EQ("string replacement matches", + text::replace_all("string without matches", + "without", "replacement")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(replace_all__several); +ATF_TEST_CASE_BODY(replace_all__several) +{ + ATF_REQUIRE_EQ("OO fOO bar OOf baz OO", + text::replace_all("oo foo bar oof baz oo", + "oo", "OO")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__ok__bool); +ATF_TEST_CASE_BODY(to_type__ok__bool) +{ + ATF_REQUIRE( text::to_type< bool >("true")); + ATF_REQUIRE(!text::to_type< bool >("false")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__ok__numerical); +ATF_TEST_CASE_BODY(to_type__ok__numerical) +{ + ATF_REQUIRE_EQ(12, text::to_type< int >("12")); + ATF_REQUIRE_EQ(18745, text::to_type< int >("18745")); + ATF_REQUIRE_EQ(-12345, text::to_type< int >("-12345")); + + ATF_REQUIRE_EQ(12.0, text::to_type< double >("12")); + ATF_REQUIRE_EQ(12.5, text::to_type< double >("12.5")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__ok__string); +ATF_TEST_CASE_BODY(to_type__ok__string) +{ + // While this seems redundant, having this particular specialization that + // does nothing allows callers to delegate work to to_type without worrying + // about the particular type being converted. + ATF_REQUIRE_EQ("", text::to_type< std::string >("")); + ATF_REQUIRE_EQ(" abcd ", text::to_type< std::string >(" abcd ")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__empty); +ATF_TEST_CASE_BODY(to_type__empty) +{ + ATF_REQUIRE_THROW(text::value_error, text::to_type< int >("")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__invalid__bool); +ATF_TEST_CASE_BODY(to_type__invalid__bool) +{ + ATF_REQUIRE_THROW(text::value_error, text::to_type< bool >("")); + ATF_REQUIRE_THROW(text::value_error, text::to_type< bool >("true ")); + ATF_REQUIRE_THROW(text::value_error, text::to_type< bool >("foo")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(to_type__invalid__numerical); +ATF_TEST_CASE_BODY(to_type__invalid__numerical) +{ + ATF_REQUIRE_THROW(text::value_error, text::to_type< int >(" 3")); + ATF_REQUIRE_THROW(text::value_error, text::to_type< int >("3 ")); + ATF_REQUIRE_THROW(text::value_error, text::to_type< int >("3a")); + ATF_REQUIRE_THROW(text::value_error, text::to_type< int >("a3")); +} + + +ATF_INIT_TEST_CASES(tcs) +{ + ATF_ADD_TEST_CASE(tcs, escape_xml__empty); + ATF_ADD_TEST_CASE(tcs, escape_xml__no_escaping); + ATF_ADD_TEST_CASE(tcs, escape_xml__some_escaping); + + ATF_ADD_TEST_CASE(tcs, quote__empty); + ATF_ADD_TEST_CASE(tcs, quote__no_escaping); + ATF_ADD_TEST_CASE(tcs, quote__some_escaping); + + ATF_ADD_TEST_CASE(tcs, refill__empty); + ATF_ADD_TEST_CASE(tcs, refill__no_changes); + ATF_ADD_TEST_CASE(tcs, refill__break_one); + ATF_ADD_TEST_CASE(tcs, refill__break_one__not_first_word); + ATF_ADD_TEST_CASE(tcs, refill__break_many); + ATF_ADD_TEST_CASE(tcs, refill__cannot_break); + ATF_ADD_TEST_CASE(tcs, refill__preserve_whitespace); + + ATF_ADD_TEST_CASE(tcs, join__empty); + ATF_ADD_TEST_CASE(tcs, join__one); + ATF_ADD_TEST_CASE(tcs, join__several); + ATF_ADD_TEST_CASE(tcs, join__unordered); + + ATF_ADD_TEST_CASE(tcs, split__empty); + ATF_ADD_TEST_CASE(tcs, split__one); + ATF_ADD_TEST_CASE(tcs, split__several__simple); + ATF_ADD_TEST_CASE(tcs, split__several__delimiters); + + ATF_ADD_TEST_CASE(tcs, replace_all__empty); + ATF_ADD_TEST_CASE(tcs, replace_all__none); + ATF_ADD_TEST_CASE(tcs, replace_all__one); + ATF_ADD_TEST_CASE(tcs, replace_all__several); + + ATF_ADD_TEST_CASE(tcs, to_type__ok__bool); + ATF_ADD_TEST_CASE(tcs, to_type__ok__numerical); + ATF_ADD_TEST_CASE(tcs, to_type__ok__string); + ATF_ADD_TEST_CASE(tcs, to_type__empty); + ATF_ADD_TEST_CASE(tcs, to_type__invalid__bool); + ATF_ADD_TEST_CASE(tcs, to_type__invalid__numerical); +} diff --git a/utils/text/regex.cpp b/utils/text/regex.cpp new file mode 100644 index 000000000000..b078ba88f6b4 --- /dev/null +++ b/utils/text/regex.cpp @@ -0,0 +1,302 @@ +// Copyright 2014 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/regex.hpp" + +extern "C" { +#include <sys/types.h> + +#include <regex.h> +} + +#include "utils/auto_array.ipp" +#include "utils/defs.hpp" +#include "utils/format/macros.hpp" +#include "utils/noncopyable.hpp" +#include "utils/sanity.hpp" +#include "utils/text/exceptions.hpp" + +namespace text = utils::text; + + +namespace { + + +static void throw_regex_error(const int, const ::regex_t*, const std::string&) + UTILS_NORETURN; + + +/// Constructs and raises a regex_error. +/// +/// \param error The error code returned by regcomp(3) or regexec(3). +/// \param preg The native regex object that caused this error. +/// \param prefix Error message prefix string. +/// +/// \throw regex_error The constructed exception. +static void +throw_regex_error(const int error, const ::regex_t* preg, + const std::string& prefix) +{ + char buffer[1024]; + + // TODO(jmmv): Would be nice to handle the case where the message does + // not fit in the temporary buffer. + (void)::regerror(error, preg, buffer, sizeof(buffer)); + + throw text::regex_error(F("%s: %s") % prefix % buffer); +} + + +} // anonymous namespace + + +/// Internal implementation for regex_matches. +struct utils::text::regex_matches::impl : utils::noncopyable { + /// String on which we are matching. + /// + /// In theory, we could take a reference here instead of a copy, and make + /// it a requirement for the caller to ensure that the lifecycle of the + /// input string outlasts the lifecycle of the regex_matches. However, that + /// contract is very easy to break with hardcoded strings (as we do in + /// tests). Just go for the safer case here. + const std::string _string; + + /// Maximum number of matching groups we expect, including the full match. + /// + /// In other words, this is the size of the _matches array. + const std::size_t _nmatches; + + /// Native regular expression match representation. + utils::auto_array< ::regmatch_t > _matches; + + /// Constructor. + /// + /// This executes the regex on the given string and sets up the internal + /// class state based on the results. + /// + /// \param preg The native regex object. + /// \param str The string on which to execute the regex. + /// \param ngroups Number of capture groups in the regex. This is an upper + /// bound and may be greater than the actual matches. + /// + /// \throw regex_error If the call to regexec(3) fails. + impl(const ::regex_t* preg, const std::string& str, + const std::size_t ngroups) : + _string(str), + _nmatches(ngroups + 1), + _matches(new ::regmatch_t[_nmatches]) + { + const int error = ::regexec(preg, _string.c_str(), _nmatches, + _matches.get(), 0); + if (error == REG_NOMATCH) { + _matches.reset(NULL); + } else if (error != 0) { + throw_regex_error(error, preg, + F("regexec on '%s' failed") % _string); + } + } + + /// Destructor. + ~impl(void) + { + } +}; + + +/// Constructor. +/// +/// \param pimpl Constructed implementation of the object. +text::regex_matches::regex_matches(std::shared_ptr< impl > pimpl) : + _pimpl(pimpl) +{ +} + + +/// Destructor. +text::regex_matches::~regex_matches(void) +{ +} + + +/// Returns the number of matches in this object. +/// +/// Note that this does not correspond to the number of groups provided at +/// construction time. The returned value here accounts for only the returned +/// valid matches. +/// +/// \return Number of matches, including the full match. +std::size_t +text::regex_matches::count(void) const +{ + std::size_t total = 0; + if (_pimpl->_matches.get() != NULL) { + for (std::size_t i = 0; i < _pimpl->_nmatches; ++i) { + if (_pimpl->_matches[i].rm_so != -1) + ++total; + } + INV(total <= _pimpl->_nmatches); + } + return total; +} + + +/// Gets a match. +/// +/// \param index Number of the match to get. Index 0 always contains the match +/// of the whole regex. +/// +/// \pre There regex must have matched the input string. +/// \pre index must be lower than count(). +/// +/// \return The textual match. +std::string +text::regex_matches::get(const std::size_t index) const +{ + PRE(*this); + PRE(index < count()); + + const ::regmatch_t* match = &_pimpl->_matches[index]; + + return std::string(_pimpl->_string.c_str() + match->rm_so, + match->rm_eo - match->rm_so); +} + + +/// Checks if there are any matches. +/// +/// \return True if the object contains one or more matches; false otherwise. +text::regex_matches::operator bool(void) const +{ + return _pimpl->_matches.get() != NULL; +} + + +/// Internal implementation for regex. +struct utils::text::regex::impl : utils::noncopyable { + /// Native regular expression representation. + ::regex_t _preg; + + /// Number of capture groups in the regular expression. This is an upper + /// bound and does NOT include the default full string match. + std::size_t _ngroups; + + /// Constructor. + /// + /// This compiles the given regular expression. + /// + /// \param regex_ The regular expression to compile. + /// \param ngroups Number of capture groups in the regular expression. This + /// is an upper bound and does NOT include the default full string + /// match. + /// \param ignore_case Whether to ignore case during matching. + /// + /// \throw regex_error If the call to regcomp(3) fails. + impl(const std::string& regex_, const std::size_t ngroups, + const bool ignore_case) : + _ngroups(ngroups) + { + const int flags = REG_EXTENDED | (ignore_case ? REG_ICASE : 0); + const int error = ::regcomp(&_preg, regex_.c_str(), flags); + if (error != 0) + throw_regex_error(error, &_preg, F("regcomp on '%s' failed") + % regex_); + } + + /// Destructor. + ~impl(void) + { + ::regfree(&_preg); + } +}; + + +/// Constructor. +/// +/// \param pimpl Constructed implementation of the object. +text::regex::regex(std::shared_ptr< impl > pimpl) : _pimpl(pimpl) +{ +} + + +/// Destructor. +text::regex::~regex(void) +{ +} + + +/// Compiles a new regular expression. +/// +/// \param regex_ The regular expression to compile. +/// \param ngroups Number of capture groups in the regular expression. This is +/// an upper bound and does NOT include the default full string match. +/// \param ignore_case Whether to ignore case during matching. +/// +/// \return A new regular expression, ready to match strings. +/// +/// \throw regex_error If the regular expression is invalid and cannot be +/// compiled. +text::regex +text::regex::compile(const std::string& regex_, const std::size_t ngroups, + const bool ignore_case) +{ + return regex(std::shared_ptr< impl >(new impl(regex_, ngroups, + ignore_case))); +} + + +/// Matches the regular expression against a string. +/// +/// \param str String to match the regular expression against. +/// +/// \return A new regex_matches object with the results of the match. +text::regex_matches +text::regex::match(const std::string& str) const +{ + std::shared_ptr< regex_matches::impl > pimpl(new regex_matches::impl( + &_pimpl->_preg, str, _pimpl->_ngroups)); + return regex_matches(pimpl); +} + + +/// Compiles and matches a regular expression once. +/// +/// This is syntactic sugar to simplify the instantiation of a new regex object +/// and its subsequent match on a string. +/// +/// \param regex_ The regular expression to compile and match. +/// \param str String to match the regular expression against. +/// \param ngroups Number of capture groups in the regular expression. +/// \param ignore_case Whether to ignore case during matching. +/// +/// \return A new regex_matches object with the results of the match. +text::regex_matches +text::match_regex(const std::string& regex_, const std::string& str, + const std::size_t ngroups, const bool ignore_case) +{ + return regex::compile(regex_, ngroups, ignore_case).match(str); +} diff --git a/utils/text/regex.hpp b/utils/text/regex.hpp new file mode 100644 index 000000000000..b3d20c246735 --- /dev/null +++ b/utils/text/regex.hpp @@ -0,0 +1,92 @@ +// Copyright 2014 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/regex.hpp +/// Utilities to build and match regular expressions. + +#if !defined(UTILS_TEXT_REGEX_HPP) +#define UTILS_TEXT_REGEX_HPP + +#include "utils/text/regex_fwd.hpp" + +#include <cstddef> +#include <memory> + + +namespace utils { +namespace text { + + +/// Container for regex match results. +class regex_matches { + struct impl; + + /// Pointer to shared implementation. + std::shared_ptr< impl > _pimpl; + + friend class regex; + regex_matches(std::shared_ptr< impl >); + +public: + ~regex_matches(void); + + std::size_t count(void) const; + std::string get(const std::size_t) const; + + operator bool(void) const; +}; + + +/// Regular expression compiler and executor. +/// +/// All regular expressions handled by this class are "extended". +class regex { + struct impl; + + /// Pointer to shared implementation. + std::shared_ptr< impl > _pimpl; + + regex(std::shared_ptr< impl >); + +public: + ~regex(void); + + static regex compile(const std::string&, const std::size_t, + const bool = false); + regex_matches match(const std::string&) const; +}; + + +regex_matches match_regex(const std::string&, const std::string&, + const std::size_t, const bool = false); + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_REGEX_HPP) diff --git a/utils/text/regex_fwd.hpp b/utils/text/regex_fwd.hpp new file mode 100644 index 000000000000..e9010324c10d --- /dev/null +++ b/utils/text/regex_fwd.hpp @@ -0,0 +1,46 @@ +// Copyright 2015 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/regex_fwd.hpp +/// Forward declarations for utils/text/regex.hpp + +#if !defined(UTILS_TEXT_REGEX_FWD_HPP) +#define UTILS_TEXT_REGEX_FWD_HPP + +namespace utils { +namespace text { + + +class regex_matches; +class regex; + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_REGEX_FWD_HPP) diff --git a/utils/text/regex_test.cpp b/utils/text/regex_test.cpp new file mode 100644 index 000000000000..7ea5ee485aad --- /dev/null +++ b/utils/text/regex_test.cpp @@ -0,0 +1,177 @@ +// Copyright 2014 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/regex.hpp" + +#include <atf-c++.hpp> + +#include "utils/text/exceptions.hpp" + +namespace text = utils::text; + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__no_matches); +ATF_TEST_CASE_BODY(integration__no_matches) +{ + const text::regex_matches matches = text::match_regex( + "foo.*bar", "this is a string without the searched text", 0); + ATF_REQUIRE(!matches); + ATF_REQUIRE_EQ(0, matches.count()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__no_capture_groups); +ATF_TEST_CASE_BODY(integration__no_capture_groups) +{ + const text::regex_matches matches = text::match_regex( + "foo.*bar", "this is a string with foo and bar embedded in it", 0); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(1, matches.count()); + ATF_REQUIRE_EQ("foo and bar", matches.get(0)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__one_capture_group); +ATF_TEST_CASE_BODY(integration__one_capture_group) +{ + const text::regex_matches matches = text::match_regex( + "^([^ ]*) ", "the string", 1); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(2, matches.count()); + ATF_REQUIRE_EQ("the ", matches.get(0)); + ATF_REQUIRE_EQ("the", matches.get(1)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__many_capture_groups); +ATF_TEST_CASE_BODY(integration__many_capture_groups) +{ + const text::regex_matches matches = text::match_regex( + "is ([^ ]*) ([a-z]*) to", "this is another string to parse", 2); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(3, matches.count()); + ATF_REQUIRE_EQ("is another string to", matches.get(0)); + ATF_REQUIRE_EQ("another", matches.get(1)); + ATF_REQUIRE_EQ("string", matches.get(2)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__capture_groups_underspecified); +ATF_TEST_CASE_BODY(integration__capture_groups_underspecified) +{ + const text::regex_matches matches = text::match_regex( + "is ([^ ]*) ([a-z]*) to", "this is another string to parse", 1); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(2, matches.count()); + ATF_REQUIRE_EQ("is another string to", matches.get(0)); + ATF_REQUIRE_EQ("another", matches.get(1)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__capture_groups_overspecified); +ATF_TEST_CASE_BODY(integration__capture_groups_overspecified) +{ + const text::regex_matches matches = text::match_regex( + "is ([^ ]*) ([a-z]*) to", "this is another string to parse", 10); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(3, matches.count()); + ATF_REQUIRE_EQ("is another string to", matches.get(0)); + ATF_REQUIRE_EQ("another", matches.get(1)); + ATF_REQUIRE_EQ("string", matches.get(2)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__reuse_regex_in_multiple_matches); +ATF_TEST_CASE_BODY(integration__reuse_regex_in_multiple_matches) +{ + const text::regex regex = text::regex::compile("number is ([0-9]+)", 1); + + { + const text::regex_matches matches = regex.match("my number is 581."); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(2, matches.count()); + ATF_REQUIRE_EQ("number is 581", matches.get(0)); + ATF_REQUIRE_EQ("581", matches.get(1)); + } + + { + const text::regex_matches matches = regex.match("your number is 6"); + ATF_REQUIRE(matches); + ATF_REQUIRE_EQ(2, matches.count()); + ATF_REQUIRE_EQ("number is 6", matches.get(0)); + ATF_REQUIRE_EQ("6", matches.get(1)); + } +} + + +ATF_TEST_CASE_WITHOUT_HEAD(integration__ignore_case); +ATF_TEST_CASE_BODY(integration__ignore_case) +{ + const text::regex regex1 = text::regex::compile("foo", 0, false); + ATF_REQUIRE(!regex1.match("bar Foo bar")); + ATF_REQUIRE(!regex1.match("bar foO bar")); + ATF_REQUIRE(!regex1.match("bar FOO bar")); + + ATF_REQUIRE(!text::match_regex("foo", "bar Foo bar", 0, false)); + ATF_REQUIRE(!text::match_regex("foo", "bar foO bar", 0, false)); + ATF_REQUIRE(!text::match_regex("foo", "bar FOO bar", 0, false)); + + const text::regex regex2 = text::regex::compile("foo", 0, true); + ATF_REQUIRE( regex2.match("bar foo bar")); + ATF_REQUIRE( regex2.match("bar Foo bar")); + ATF_REQUIRE( regex2.match("bar foO bar")); + ATF_REQUIRE( regex2.match("bar FOO bar")); + + ATF_REQUIRE( text::match_regex("foo", "bar foo bar", 0, true)); + ATF_REQUIRE( text::match_regex("foo", "bar Foo bar", 0, true)); + ATF_REQUIRE( text::match_regex("foo", "bar foO bar", 0, true)); + ATF_REQUIRE( text::match_regex("foo", "bar FOO bar", 0, true)); +} + +ATF_TEST_CASE_WITHOUT_HEAD(integration__invalid_regex); +ATF_TEST_CASE_BODY(integration__invalid_regex) +{ + ATF_REQUIRE_THROW(text::regex_error, + text::regex::compile("this is (unbalanced", 0)); +} + + +ATF_INIT_TEST_CASES(tcs) +{ + // regex and regex_matches are so coupled that it makes no sense to test + // them independently. Just validate their integration. + ATF_ADD_TEST_CASE(tcs, integration__no_matches); + ATF_ADD_TEST_CASE(tcs, integration__no_capture_groups); + ATF_ADD_TEST_CASE(tcs, integration__one_capture_group); + ATF_ADD_TEST_CASE(tcs, integration__many_capture_groups); + ATF_ADD_TEST_CASE(tcs, integration__capture_groups_underspecified); + ATF_ADD_TEST_CASE(tcs, integration__capture_groups_overspecified); + ATF_ADD_TEST_CASE(tcs, integration__reuse_regex_in_multiple_matches); + ATF_ADD_TEST_CASE(tcs, integration__ignore_case); + ATF_ADD_TEST_CASE(tcs, integration__invalid_regex); +} diff --git a/utils/text/table.cpp b/utils/text/table.cpp new file mode 100644 index 000000000000..4a2c72f8053f --- /dev/null +++ b/utils/text/table.cpp @@ -0,0 +1,428 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/table.hpp" + +#include <algorithm> +#include <iterator> +#include <limits> +#include <sstream> + +#include "utils/sanity.hpp" +#include "utils/text/operations.ipp" + +namespace text = utils::text; + + +namespace { + + +/// Applies user overrides to the column widths of a table. +/// +/// \param table The table from which to calculate the column widths. +/// \param user_widths The column widths provided by the user. This vector must +/// have less or the same number of elements as the columns of the table. +/// Values of width_auto are ignored; any other explicit values are copied +/// to the output widths vector, including width_refill. +/// +/// \return A vector with the widths of the columns of the input table with any +/// user overrides applied. +static text::widths_vector +override_column_widths(const text::table& table, + const text::widths_vector& user_widths) +{ + PRE(user_widths.size() <= table.ncolumns()); + text::widths_vector widths = table.column_widths(); + + // Override the actual width of the columns based on user-specified widths. + for (text::widths_vector::size_type i = 0; i < user_widths.size(); ++i) { + const text::widths_vector::value_type& user_width = user_widths[i]; + if (user_width != text::table_formatter::width_auto) { + PRE_MSG(user_width == text::table_formatter::width_refill || + user_width >= widths[i], + "User-provided column widths must be larger than the " + "column contents (except for the width_refill column)"); + widths[i] = user_width; + } + } + + return widths; +} + + +/// Locates the refill column, if any. +/// +/// \param widths The widths of the columns as returned by +/// override_column_widths(). Note that one of the columns may or may not +/// be width_refill, which is the column we are looking for. +/// +/// \return The index of the refill column with a width_refill width if any, or +/// otherwise the index of the last column (which is the default refill column). +static text::widths_vector::size_type +find_refill_column(const text::widths_vector& widths) +{ + text::widths_vector::size_type i = 0; + for (; i < widths.size(); ++i) { + if (widths[i] == text::table_formatter::width_refill) + return i; + } + return i - 1; +} + + +/// Pads the widths of the table to fit within a maximum width. +/// +/// On output, a column of the widths vector is truncated to a shorter length +/// than its current value, if the total width of the table would exceed the +/// maximum table width. +/// +/// \param [in,out] widths The widths of the columns as returned by +/// override_column_widths(). One of these columns should have a value of +/// width_refill; if not, a default column is refilled. +/// \param user_max_width The target width of the table; must not be zero. +/// \param column_padding The padding between the cells, if any. The target +/// width should be larger than the padding times the number of columns; if +/// that is not the case, we attempt a readjustment here. +static void +refill_widths(text::widths_vector& widths, + const text::widths_vector::value_type user_max_width, + const std::size_t column_padding) +{ + PRE(user_max_width != 0); + + // widths.size() is a proxy for the number of columns of the table. + const std::size_t total_padding = column_padding * (widths.size() - 1); + const text::widths_vector::value_type max_width = std::max( + user_max_width, total_padding) - total_padding; + + const text::widths_vector::size_type refill_column = + find_refill_column(widths); + INV(refill_column < widths.size()); + + text::widths_vector::value_type width = 0; + for (text::widths_vector::size_type i = 0; i < widths.size(); ++i) { + if (i != refill_column) + width += widths[i]; + } + widths[refill_column] = max_width - width; +} + + +/// Pads an input text to a specified width with spaces. +/// +/// \param input The text to add padding to (may be empty). +/// \param length The desired length of the output. +/// \param is_last Whether the text being processed belongs to the last column +/// of a row or not. Values in the last column should not be padded to +/// prevent trailing whitespace on the screen (which affects copy/pasting +/// for example). +/// +/// \return The padded cell. If the input string is longer than the desired +/// length, the input string is returned verbatim. The padded table won't be +/// correct, but we don't expect this to be a common case to worry about. +static std::string +pad_cell(const std::string& input, const std::size_t length, const bool is_last) +{ + if (is_last) + return input; + else { + if (input.length() < length) + return input + std::string(length - input.length(), ' '); + else + return input; + } +} + + +/// Refills a cell and adds it to the output lines. +/// +/// \param row The row containing the cell to be refilled. +/// \param widths The widths of the row. +/// \param column The column being refilled. +/// \param [in,out] textual_rows The output lines as processed so far. This is +/// updated to accomodate for the contents of the refilled cell, extending +/// the rows as necessary. +static void +refill_cell(const text::table_row& row, const text::widths_vector& widths, + const text::table_row::size_type column, + std::vector< text::table_row >& textual_rows) +{ + const std::vector< std::string > rows = text::refill(row[column], + widths[column]); + + if (textual_rows.size() < rows.size()) + textual_rows.resize(rows.size(), text::table_row(row.size())); + + for (std::vector< std::string >::size_type i = 0; i < rows.size(); ++i) { + for (text::table_row::size_type j = 0; j < row.size(); ++j) { + const bool is_last = j == row.size() - 1; + if (j == column) + textual_rows[i][j] = pad_cell(rows[i], widths[j], is_last); + else { + if (textual_rows[i][j].empty()) + textual_rows[i][j] = pad_cell("", widths[j], is_last); + } + } + } +} + + +/// Formats a single table row. +/// +/// \param row The row to format. +/// \param widths The widths of the columns to apply during formatting. Cells +/// wider than the specified width are refilled to attempt to fit in the +/// cell. Cells narrower than the width are right-padded with spaces. +/// \param separator The column separator to use. +/// +/// \return The textual lines that contain the formatted row. +static std::vector< std::string > +format_row(const text::table_row& row, const text::widths_vector& widths, + const std::string& separator) +{ + PRE(row.size() == widths.size()); + + std::vector< text::table_row > textual_rows(1, text::table_row(row.size())); + + for (text::table_row::size_type column = 0; column < row.size(); ++column) { + if (widths[column] > row[column].length()) + textual_rows[0][column] = pad_cell(row[column], widths[column], + column == row.size() - 1); + else + refill_cell(row, widths, column, textual_rows); + } + + std::vector< std::string > lines; + for (std::vector< text::table_row >::const_iterator + iter = textual_rows.begin(); iter != textual_rows.end(); ++iter) { + lines.push_back(text::join(*iter, separator)); + } + return lines; +} + + +} // anonymous namespace + + +/// Constructs a new table. +/// +/// \param ncolumns_ The number of columns that the table will have. +text::table::table(const table_row::size_type ncolumns_) +{ + _column_widths.resize(ncolumns_, 0); +} + + +/// Gets the number of columns in the table. +/// +/// \return The number of columns in the table. This value remains constant +/// during the existence of the table. +text::widths_vector::size_type +text::table::ncolumns(void) const +{ + return _column_widths.size(); +} + + +/// Gets the width of a column. +/// +/// The returned value is not valid if add_row() is called again, as the column +/// may have grown in width. +/// +/// \param column The index of the column of which to get the width. Must be +/// less than the total number of columns. +/// +/// \return The width of a column. +text::widths_vector::value_type +text::table::column_width(const widths_vector::size_type column) const +{ + PRE(column < _column_widths.size()); + return _column_widths[column]; +} + + +/// Gets the widths of all columns. +/// +/// The returned value is not valid if add_row() is called again, as the columns +/// may have grown in width. +/// +/// \return A vector with the width of all columns. +const text::widths_vector& +text::table::column_widths(void) const +{ + return _column_widths; +} + + +/// Checks whether the table is empty or not. +/// +/// \return True if the table is empty; false otherwise. +bool +text::table::empty(void) const +{ + return _rows.empty(); +} + + +/// Adds a row to the table. +/// +/// \param row The row to be added. This row must have the same amount of +/// columns as defined during the construction of the table. +void +text::table::add_row(const table_row& row) +{ + PRE(row.size() == _column_widths.size()); + _rows.push_back(row); + + for (table_row::size_type i = 0; i < row.size(); ++i) + if (_column_widths[i] < row[i].length()) + _column_widths[i] = row[i].length(); +} + + +/// Gets an iterator pointing to the beginning of the rows of the table. +/// +/// \return An iterator on the rows. +text::table::const_iterator +text::table::begin(void) const +{ + return _rows.begin(); +} + + +/// Gets an iterator pointing to the end of the rows of the table. +/// +/// \return An iterator on the rows. +text::table::const_iterator +text::table::end(void) const +{ + return _rows.end(); +} + + +/// Column width to denote that the column has to fit all of its cells. +const std::size_t text::table_formatter::width_auto = 0; + + +/// Column width to denote that the column can be refilled to fit the table. +const std::size_t text::table_formatter::width_refill = + std::numeric_limits< std::size_t >::max(); + + +/// Constructs a new table formatter. +text::table_formatter::table_formatter(void) : + _separator(""), + _table_width(0) +{ +} + + +/// Sets the width of a column. +/// +/// All columns except one must have a width that is, at least, as wide as the +/// widest cell in the column. One of the columns can have a width of +/// width_refill, which indicates that the column will be refilled if the table +/// does not fit in its maximum width. +/// +/// \param column The index of the column to set the width for. +/// \param width The width to set the column to. +/// +/// \return A reference to this formatter to allow using the builder pattern. +text::table_formatter& +text::table_formatter::set_column_width(const table_row::size_type column, + const std::size_t width) +{ +#if !defined(NDEBUG) + if (width == width_refill) { + for (widths_vector::size_type i = 0; i < _column_widths.size(); i++) { + if (i != column) + PRE_MSG(_column_widths[i] != width_refill, + "Only one column width can be set to width_refill"); + } + } +#endif + + if (_column_widths.size() < column + 1) + _column_widths.resize(column + 1, width_auto); + _column_widths[column] = width; + return *this; +} + + +/// Sets the separator to use between the cells. +/// +/// \param separator The separator to use. +/// +/// \return A reference to this formatter to allow using the builder pattern. +text::table_formatter& +text::table_formatter::set_separator(const char* separator) +{ + _separator = separator; + return *this; +} + + +/// Sets the maximum width of the table. +/// +/// \param table_width The maximum width of the table; cannot be zero. +/// +/// \return A reference to this formatter to allow using the builder pattern. +text::table_formatter& +text::table_formatter::set_table_width(const std::size_t table_width) +{ + PRE(table_width > 0); + _table_width = table_width; + return *this; +} + + +/// Formats a table into a collection of textual lines. +/// +/// \param t Table to format. +/// +/// \return A collection of textual lines. +std::vector< std::string > +text::table_formatter::format(const table& t) const +{ + std::vector< std::string > lines; + + if (!t.empty()) { + widths_vector widths = override_column_widths(t, _column_widths); + if (_table_width != 0) + refill_widths(widths, _table_width, _separator.length()); + + for (table::const_iterator iter = t.begin(); iter != t.end(); ++iter) { + const std::vector< std::string > sublines = + format_row(*iter, widths, _separator); + std::copy(sublines.begin(), sublines.end(), + std::back_inserter(lines)); + } + } + + return lines; +} diff --git a/utils/text/table.hpp b/utils/text/table.hpp new file mode 100644 index 000000000000..5fd7c50c991c --- /dev/null +++ b/utils/text/table.hpp @@ -0,0 +1,125 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/table.hpp +/// Table construction and formatting. + +#if !defined(UTILS_TEXT_TABLE_HPP) +#define UTILS_TEXT_TABLE_HPP + +#include "utils/text/table_fwd.hpp" + +#include <cstddef> +#include <string> +#include <vector> + +namespace utils { +namespace text { + + +/// Representation of a table. +/// +/// A table is nothing more than a matrix of rows by columns. The number of +/// columns is hardcoded at construction times, and the rows can be accumulated +/// at a later stage. +/// +/// The only value of this class is a simpler and more natural mechanism of the +/// construction of a table, with additional sanity checks. We could as well +/// just expose the internal data representation to our users. +class table { + /// Widths of the table columns so far. + widths_vector _column_widths; + + /// Type defining the collection of rows in the table. + typedef std::vector< table_row > rows_vector; + + /// The rows of the table. + /// + /// This is actually the matrix representing the table. Every element of + /// this vector (which are vectors themselves) must have _ncolumns items. + rows_vector _rows; + +public: + table(const table_row::size_type); + + widths_vector::size_type ncolumns(void) const; + widths_vector::value_type column_width(const widths_vector::size_type) + const; + const widths_vector& column_widths(void) const; + + void add_row(const table_row&); + + bool empty(void) const; + + /// Constant iterator on the rows of the table. + typedef rows_vector::const_iterator const_iterator; + + const_iterator begin(void) const; + const_iterator end(void) const; +}; + + +/// Settings to format a table. +/// +/// This class implements a builder pattern to construct an object that contains +/// all the knowledge to format a table. Once all the settings have been set, +/// the format() method provides the algorithm to apply such formatting settings +/// to any input table. +class table_formatter { + /// Text to use as the separator between cells. + std::string _separator; + + /// Colletion of widths of the columns of a table. + std::size_t _table_width; + + /// Widths of the table columns. + /// + /// Note that this only includes widths for the column widths explicitly + /// overriden by the caller. In other words, this vector can be shorter + /// than the table passed to the format() method, which is just fine. Any + /// non-specified column widths are assumed to be width_auto. + widths_vector _column_widths; + +public: + table_formatter(void); + + static const std::size_t width_auto; + static const std::size_t width_refill; + table_formatter& set_column_width(const table_row::size_type, + const std::size_t); + table_formatter& set_separator(const char*); + table_formatter& set_table_width(const std::size_t); + + std::vector< std::string > format(const table&) const; +}; + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_TABLE_HPP) diff --git a/utils/text/table_fwd.hpp b/utils/text/table_fwd.hpp new file mode 100644 index 000000000000..77c6b1fa8c78 --- /dev/null +++ b/utils/text/table_fwd.hpp @@ -0,0 +1,58 @@ +// Copyright 2015 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/table_fwd.hpp +/// Forward declarations for utils/text/table.hpp + +#if !defined(UTILS_TEXT_TABLE_FWD_HPP) +#define UTILS_TEXT_TABLE_FWD_HPP + +#include <cstddef> +#include <string> +#include <vector> + +namespace utils { +namespace text { + + +/// Values of the cells of a particular table row. +typedef std::vector< std::string > table_row; + + +/// Vector of column widths. +typedef std::vector< std::size_t > widths_vector; + + +class table; +class table_formatter; + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_TABLE_FWD_HPP) diff --git a/utils/text/table_test.cpp b/utils/text/table_test.cpp new file mode 100644 index 000000000000..45928dae89c4 --- /dev/null +++ b/utils/text/table_test.cpp @@ -0,0 +1,413 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/table.hpp" + +#include <algorithm> + +#include <atf-c++.hpp> + +#include "utils/text/operations.ipp" + +namespace text = utils::text; + + +/// Performs a check on text::table_formatter. +/// +/// This is provided for test simplicity's sake. Having to match the result of +/// the formatting on a line by line basis would result in too verbose tests +/// (maybe not with C++11, but not using this yet). +/// +/// Because of the flattening of the formatted table into a string, we risk +/// misdetecting problems when the algorithm bundles newlines into the lines of +/// a table. This should not happen, and not accounting for this little detail +/// makes testing so much easier. +/// +/// \param expected Textual representation of the table, as a collection of +/// lines separated by newline characters. +/// \param formatter The formatter to use. +/// \param table The table to format. +static void +table_formatter_check(const std::string& expected, + const text::table_formatter& formatter, + const text::table& table) +{ + ATF_REQUIRE_EQ(expected, text::join(formatter.format(table), "\n") + "\n"); +} + + + +ATF_TEST_CASE_WITHOUT_HEAD(table__ncolumns); +ATF_TEST_CASE_BODY(table__ncolumns) +{ + ATF_REQUIRE_EQ(5, text::table(5).ncolumns()); + ATF_REQUIRE_EQ(10, text::table(10).ncolumns()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table__column_width); +ATF_TEST_CASE_BODY(table__column_width) +{ + text::table_row row1; + row1.push_back("1234"); + row1.push_back("123456"); + text::table_row row2; + row2.push_back("12"); + row2.push_back("12345678"); + + text::table table(2); + table.add_row(row1); + table.add_row(row2); + + ATF_REQUIRE_EQ(4, table.column_width(0)); + ATF_REQUIRE_EQ(8, table.column_width(1)); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table__column_widths); +ATF_TEST_CASE_BODY(table__column_widths) +{ + text::table_row row1; + row1.push_back("1234"); + row1.push_back("123456"); + text::table_row row2; + row2.push_back("12"); + row2.push_back("12345678"); + + text::table table(2); + table.add_row(row1); + table.add_row(row2); + + ATF_REQUIRE_EQ(4, table.column_widths()[0]); + ATF_REQUIRE_EQ(8, table.column_widths()[1]); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table__empty); +ATF_TEST_CASE_BODY(table__empty) +{ + text::table table(2); + ATF_REQUIRE(table.empty()); + table.add_row(text::table_row(2)); + ATF_REQUIRE(!table.empty()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table__iterate); +ATF_TEST_CASE_BODY(table__iterate) +{ + text::table_row row1; + row1.push_back("foo"); + text::table_row row2; + row2.push_back("bar"); + + text::table table(1); + table.add_row(row1); + table.add_row(row2); + + text::table::const_iterator iter = table.begin(); + ATF_REQUIRE(iter != table.end()); + ATF_REQUIRE(row1 == *iter); + ++iter; + ATF_REQUIRE(iter != table.end()); + ATF_REQUIRE(row2 == *iter); + ++iter; + ATF_REQUIRE(iter == table.end()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__empty); +ATF_TEST_CASE_BODY(table_formatter__empty) +{ + ATF_REQUIRE(text::table_formatter().set_separator(" ") + .format(text::table(1)).empty()); + ATF_REQUIRE(text::table_formatter().set_separator(" ") + .format(text::table(10)).empty()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__defaults); +ATF_TEST_CASE_BODY(table_formatter__defaults) +{ + text::table table(3); + { + text::table_row row; + row.push_back("First"); + row.push_back("Second"); + row.push_back("Third"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Fourth with some text"); + row.push_back("Fifth with some more text"); + row.push_back("Sixth foo"); + table.add_row(row); + } + + table_formatter_check( + "First Second Third\n" + "Fourth with some textFifth with some more textSixth foo\n", + text::table_formatter(), table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__one_column__no_max_width); +ATF_TEST_CASE_BODY(table_formatter__one_column__no_max_width) +{ + text::table table(1); + { + text::table_row row; + row.push_back("First row with some words"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Second row with some words"); + table.add_row(row); + } + + table_formatter_check( + "First row with some words\n" + "Second row with some words\n", + text::table_formatter().set_separator(" | "), table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__one_column__explicit_width); +ATF_TEST_CASE_BODY(table_formatter__one_column__explicit_width) +{ + text::table table(1); + { + text::table_row row; + row.push_back("First row with some words"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Second row with some words"); + table.add_row(row); + } + + table_formatter_check( + "First row with some words\n" + "Second row with some words\n", + text::table_formatter().set_separator(" | ").set_column_width(0, 1024), + table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__one_column__max_width); +ATF_TEST_CASE_BODY(table_formatter__one_column__max_width) +{ + text::table table(1); + { + text::table_row row; + row.push_back("First row with some words"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Second row with some words"); + table.add_row(row); + } + + table_formatter_check( + "First row\nwith some\nwords\n" + "Second row\nwith some\nwords\n", + text::table_formatter().set_separator(" | ").set_table_width(11), + table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__many_columns__no_max_width); +ATF_TEST_CASE_BODY(table_formatter__many_columns__no_max_width) +{ + text::table table(3); + { + text::table_row row; + row.push_back("First"); + row.push_back("Second"); + row.push_back("Third"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Fourth with some text"); + row.push_back("Fifth with some more text"); + row.push_back("Sixth foo"); + table.add_row(row); + } + + table_formatter_check( + "First | Second | Third\n" + "Fourth with some text | Fifth with some more text | Sixth foo\n", + text::table_formatter().set_separator(" | "), table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__many_columns__explicit_width); +ATF_TEST_CASE_BODY(table_formatter__many_columns__explicit_width) +{ + text::table table(3); + { + text::table_row row; + row.push_back("First"); + row.push_back("Second"); + row.push_back("Third"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Fourth with some text"); + row.push_back("Fifth with some more text"); + row.push_back("Sixth foo"); + table.add_row(row); + } + + table_formatter_check( + "First | Second | Third\n" + "Fourth with some text | Fifth with some more text | Sixth foo\n", + text::table_formatter().set_separator(" | ").set_column_width(0, 23) + .set_column_width(1, 28), table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__many_columns__max_width); +ATF_TEST_CASE_BODY(table_formatter__many_columns__max_width) +{ + text::table table(3); + { + text::table_row row; + row.push_back("First"); + row.push_back("Second"); + row.push_back("Third"); + table.add_row(row); + } + { + text::table_row row; + row.push_back("Fourth with some text"); + row.push_back("Fifth with some more text"); + row.push_back("Sixth foo"); + table.add_row(row); + } + + table_formatter_check( + "First | Second | Third\n" + "Fourth with some text | Fifth with | Sixth foo\n" + " | some more | \n" + " | text | \n", + text::table_formatter().set_separator(" | ").set_table_width(46) + .set_column_width(1, text::table_formatter::width_refill) + .set_column_width(0, text::table_formatter::width_auto), table); + + table_formatter_check( + "First | Second | Third\n" + "Fourth with some text | Fifth with | Sixth foo\n" + " | some more | \n" + " | text | \n", + text::table_formatter().set_separator(" | ").set_table_width(48) + .set_column_width(1, text::table_formatter::width_refill) + .set_column_width(0, 23), table); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(table_formatter__use_case__cli_help); +ATF_TEST_CASE_BODY(table_formatter__use_case__cli_help) +{ + text::table options_table(2); + { + text::table_row row; + row.push_back("-a a_value"); + row.push_back("This is the description of the first flag"); + options_table.add_row(row); + } + { + text::table_row row; + row.push_back("-b"); + row.push_back("And this is the text for the second flag"); + options_table.add_row(row); + } + + text::table commands_table(2); + { + text::table_row row; + row.push_back("first"); + row.push_back("This is the first command"); + commands_table.add_row(row); + } + { + text::table_row row; + row.push_back("second"); + row.push_back("And this is the second command"); + commands_table.add_row(row); + } + + const text::widths_vector::value_type first_width = + std::max(options_table.column_width(0), commands_table.column_width(0)); + + table_formatter_check( + "-a a_value This is the description\n" + " of the first flag\n" + "-b And this is the text for\n" + " the second flag\n", + text::table_formatter().set_separator(" ").set_table_width(36) + .set_column_width(0, first_width) + .set_column_width(1, text::table_formatter::width_refill), + options_table); + + table_formatter_check( + "first This is the first\n" + " command\n" + "second And this is the second\n" + " command\n", + text::table_formatter().set_separator(" ").set_table_width(36) + .set_column_width(0, first_width) + .set_column_width(1, text::table_formatter::width_refill), + commands_table); +} + + +ATF_INIT_TEST_CASES(tcs) +{ + ATF_ADD_TEST_CASE(tcs, table__ncolumns); + ATF_ADD_TEST_CASE(tcs, table__column_width); + ATF_ADD_TEST_CASE(tcs, table__column_widths); + ATF_ADD_TEST_CASE(tcs, table__empty); + ATF_ADD_TEST_CASE(tcs, table__iterate); + + ATF_ADD_TEST_CASE(tcs, table_formatter__empty); + ATF_ADD_TEST_CASE(tcs, table_formatter__defaults); + ATF_ADD_TEST_CASE(tcs, table_formatter__one_column__no_max_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__one_column__explicit_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__one_column__max_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__many_columns__no_max_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__many_columns__explicit_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__many_columns__max_width); + ATF_ADD_TEST_CASE(tcs, table_formatter__use_case__cli_help); +} diff --git a/utils/text/templates.cpp b/utils/text/templates.cpp new file mode 100644 index 000000000000..13cb27b1cce2 --- /dev/null +++ b/utils/text/templates.cpp @@ -0,0 +1,764 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/templates.hpp" + +#include <algorithm> +#include <fstream> +#include <sstream> +#include <stack> + +#include "utils/format/macros.hpp" +#include "utils/fs/path.hpp" +#include "utils/noncopyable.hpp" +#include "utils/sanity.hpp" +#include "utils/text/exceptions.hpp" +#include "utils/text/operations.ipp" + +namespace text = utils::text; + + +namespace { + + +/// Definition of a template statement. +/// +/// A template statement is a particular line in the input file that is +/// preceeded by a template marker. This class provides a high-level +/// representation of the contents of such statement and a mechanism to parse +/// the textual line into this high-level representation. +class statement_def { +public: + /// Types of the known statements. + enum statement_type { + /// Alternative clause of a conditional. + /// + /// Takes no arguments. + type_else, + + /// End of conditional marker. + /// + /// Takes no arguments. + type_endif, + + /// End of loop marker. + /// + /// Takes no arguments. + type_endloop, + + /// Beginning of a conditional. + /// + /// Takes a single argument, which denotes the name of the variable or + /// vector to check for existence. This is the only expression + /// supported. + type_if, + + /// Beginning of a loop over all the elements of a vector. + /// + /// Takes two arguments: the name of the vector over which to iterate + /// and the name of the iterator to later index this vector. + type_loop, + }; + +private: + /// Internal data describing the structure of a particular statement type. + struct type_descriptor { + /// The native type of the statement. + statement_type type; + + /// The expected number of arguments. + unsigned int n_arguments; + + /// Constructs a new type descriptor. + /// + /// \param type_ The native type of the statement. + /// \param n_arguments_ The expected number of arguments. + type_descriptor(const statement_type type_, + const unsigned int n_arguments_) + : type(type_), n_arguments(n_arguments_) + { + } + }; + + /// Mapping of statement type names to their definitions. + typedef std::map< std::string, type_descriptor > types_map; + + /// Description of the different statement types. + /// + /// This static map is initialized once and reused later for any statement + /// lookup. Unfortunately, we cannot perform this initialization in a + /// static manner without C++11. + static types_map _types; + + /// Generates a new types definition map. + /// + /// \return A new types definition map, to be assigned to _types. + static types_map + generate_types_map(void) + { + // If you change this, please edit the comments in the enum above. + types_map types; + types.insert(types_map::value_type( + "else", type_descriptor(type_else, 0))); + types.insert(types_map::value_type( + "endif", type_descriptor(type_endif, 0))); + types.insert(types_map::value_type( + "endloop", type_descriptor(type_endloop, 0))); + types.insert(types_map::value_type( + "if", type_descriptor(type_if, 1))); + types.insert(types_map::value_type( + "loop", type_descriptor(type_loop, 2))); + return types; + } + +public: + /// The type of the statement. + statement_type type; + + /// The arguments to the statement, in textual form. + const std::vector< std::string > arguments; + + /// Creates a new statement. + /// + /// \param type_ The type of the statement. + /// \param arguments_ The arguments to the statement. + statement_def(const statement_type& type_, + const std::vector< std::string >& arguments_) : + type(type_), arguments(arguments_) + { +#if !defined(NDEBUG) + for (types_map::const_iterator iter = _types.begin(); + iter != _types.end(); ++iter) { + const type_descriptor& descriptor = (*iter).second; + if (descriptor.type == type_) { + PRE(descriptor.n_arguments == arguments_.size()); + return; + } + } + UNREACHABLE; +#endif + } + + /// Parses a statement. + /// + /// \param line The textual representation of the statement without any + /// prefix. + /// + /// \return The parsed statement. + /// + /// \throw text::syntax_error If the statement is not correctly defined. + static statement_def + parse(const std::string& line) + { + if (_types.empty()) + _types = generate_types_map(); + + const std::vector< std::string > words = text::split(line, ' '); + if (words.empty()) + throw text::syntax_error("Empty statement"); + + const types_map::const_iterator iter = _types.find(words[0]); + if (iter == _types.end()) + throw text::syntax_error(F("Unknown statement '%s'") % words[0]); + const type_descriptor& descriptor = (*iter).second; + + if (words.size() - 1 != descriptor.n_arguments) + throw text::syntax_error(F("Invalid number of arguments for " + "statement '%s'") % words[0]); + + std::vector< std::string > new_arguments; + new_arguments.resize(words.size() - 1); + std::copy(words.begin() + 1, words.end(), new_arguments.begin()); + + return statement_def(descriptor.type, new_arguments); + } +}; + + +statement_def::types_map statement_def::_types; + + +/// Definition of a loop. +/// +/// This simple structure is used to keep track of the parameters of a loop. +struct loop_def { + /// The name of the vector over which this loop is iterating. + std::string vector; + + /// The name of the iterator defined by this loop. + std::string iterator; + + /// Position in the input to which to rewind to on looping. + /// + /// This position points to the line after the loop statement, not the loop + /// itself. This is one of the reasons why we have this structure, so that + /// we can maintain the data about the loop without having to re-process it. + std::istream::pos_type position; + + /// Constructs a new loop definition. + /// + /// \param vector_ The name of the vector (first argument). + /// \param iterator_ The name of the iterator (second argumnet). + /// \param position_ Position of the next line after the loop statement. + loop_def(const std::string& vector_, const std::string& iterator_, + const std::istream::pos_type position_) : + vector(vector_), iterator(iterator_), position(position_) + { + } +}; + + +/// Stateful class to instantiate the templates in an input stream. +/// +/// The goal of this parser is to scan the input once and not buffer anything in +/// memory. The only exception are loops: loops are reinterpreted on every +/// iteration from the same input file by rewidining the stream to the +/// appropriate position. +class templates_parser : utils::noncopyable { + /// The templates to apply. + /// + /// Note that this is not const because the parser has to have write access + /// to the templates. In particular, it needs to be able to define the + /// iterators as regular variables. + text::templates_def _templates; + + /// Prefix that marks a line as a statement. + const std::string _prefix; + + /// Delimiter to surround an expression instantiation. + const std::string _delimiter; + + /// Whether to skip incoming lines or not. + /// + /// The top of the stack is true whenever we encounter a conditional that + /// evaluates to false or a loop that does not have any iterations left. + /// Under these circumstances, we need to continue scanning the input stream + /// until we find the matching closing endif or endloop construct. + /// + /// This is a stack rather than a plain boolean to allow us deal with + /// if-else clauses. + std::stack< bool > _skip; + + /// Current count of nested conditionals. + unsigned int _if_level; + + /// Level of the top-most conditional that evaluated to false. + unsigned int _exit_if_level; + + /// Current count of nested loops. + unsigned int _loop_level; + + /// Level of the top-most loop that does not have any iterations left. + unsigned int _exit_loop_level; + + /// Information about all the nested loops up to the current point. + std::stack< loop_def > _loops; + + /// Checks if a line is a statement or not. + /// + /// \param line The line to validate. + /// + /// \return True if the line looks like a statement, which is determined by + /// checking if the line starts by the predefined prefix. + bool + is_statement(const std::string& line) + { + return ((line.length() >= _prefix.length() && + line.substr(0, _prefix.length()) == _prefix) && + (line.length() < _delimiter.length() || + line.substr(0, _delimiter.length()) != _delimiter)); + } + + /// Parses a given statement line into a statement definition. + /// + /// \param line The line to validate; it must be a valid statement. + /// + /// \return The parsed statement. + /// + /// \throw text::syntax_error If the input is not a valid statement. + statement_def + parse_statement(const std::string& line) + { + PRE(is_statement(line)); + return statement_def::parse(line.substr(_prefix.length())); + } + + /// Processes a line from the input when not in skip mode. + /// + /// \param line The line to be processed. + /// \param input The input stream from which the line was read. The current + /// position in the stream must be after the line being processed. + /// \param output The output stream into which to write the results. + /// + /// \throw text::syntax_error If the input is not valid. + void + handle_normal(const std::string& line, std::istream& input, + std::ostream& output) + { + if (!is_statement(line)) { + // Fast path. Mostly to avoid an indentation level for the big + // chunk of code below. + output << line << '\n'; + return; + } + + const statement_def statement = parse_statement(line); + + switch (statement.type) { + case statement_def::type_else: + _skip.top() = !_skip.top(); + break; + + case statement_def::type_endif: + _if_level--; + break; + + case statement_def::type_endloop: { + PRE(_loops.size() == _loop_level); + loop_def& loop = _loops.top(); + + const std::size_t next_index = 1 + text::to_type< std::size_t >( + _templates.get_variable(loop.iterator)); + + if (next_index < _templates.get_vector(loop.vector).size()) { + _templates.add_variable(loop.iterator, F("%s") % next_index); + input.seekg(loop.position); + } else { + _loop_level--; + _loops.pop(); + _templates.remove_variable(loop.iterator); + } + } break; + + case statement_def::type_if: { + _if_level++; + const std::string value = _templates.evaluate( + statement.arguments[0]); + if (value.empty() || value == "0" || value == "false") { + _exit_if_level = _if_level; + _skip.push(true); + } else { + _skip.push(false); + } + } break; + + case statement_def::type_loop: { + _loop_level++; + + const loop_def loop(statement.arguments[0], statement.arguments[1], + input.tellg()); + if (_templates.get_vector(loop.vector).empty()) { + _exit_loop_level = _loop_level; + _skip.push(true); + } else { + _templates.add_variable(loop.iterator, "0"); + _loops.push(loop); + _skip.push(false); + } + } break; + } + } + + /// Processes a line from the input when in skip mode. + /// + /// \param line The line to be processed. + /// + /// \throw text::syntax_error If the input is not valid. + void + handle_skip(const std::string& line) + { + PRE(_skip.top()); + + if (!is_statement(line)) + return; + + const statement_def statement = parse_statement(line); + switch (statement.type) { + case statement_def::type_else: + if (_exit_if_level == _if_level) + _skip.top() = !_skip.top(); + break; + + case statement_def::type_endif: + INV(_if_level >= _exit_if_level); + if (_if_level == _exit_if_level) + _skip.top() = false; + _if_level--; + _skip.pop(); + break; + + case statement_def::type_endloop: + INV(_loop_level >= _exit_loop_level); + if (_loop_level == _exit_loop_level) + _skip.top() = false; + _loop_level--; + _skip.pop(); + break; + + case statement_def::type_if: + _if_level++; + _skip.push(true); + break; + + case statement_def::type_loop: + _loop_level++; + _skip.push(true); + break; + + default: + break; + } + } + + /// Evaluates expressions on a given input line. + /// + /// An expression is surrounded by _delimiter on both sides. We scan the + /// string from left to right finding any expressions that may appear, yank + /// them out and call templates_def::evaluate() to get their value. + /// + /// Lonely or unbalanced appearances of _delimiter on the input line are + /// not considered an error, given that the user may actually want to supply + /// that character sequence without being interpreted as a template. + /// + /// \param in_line The input line from which to evaluate expressions. + /// + /// \return The evaluated line. + /// + /// \throw text::syntax_error If the expressions in the line are malformed. + std::string + evaluate(const std::string& in_line) + { + std::string out_line; + + std::string::size_type last_pos = 0; + while (last_pos != std::string::npos) { + const std::string::size_type open_pos = in_line.find( + _delimiter, last_pos); + if (open_pos == std::string::npos) { + out_line += in_line.substr(last_pos); + last_pos = std::string::npos; + } else { + const std::string::size_type close_pos = in_line.find( + _delimiter, open_pos + _delimiter.length()); + if (close_pos == std::string::npos) { + out_line += in_line.substr(last_pos); + last_pos = std::string::npos; + } else { + out_line += in_line.substr(last_pos, open_pos - last_pos); + out_line += _templates.evaluate(in_line.substr( + open_pos + _delimiter.length(), + close_pos - open_pos - _delimiter.length())); + last_pos = close_pos + _delimiter.length(); + } + } + } + + return out_line; + } + +public: + /// Constructs a new template parser. + /// + /// \param templates_ The templates to apply to the processed file. + /// \param prefix_ The prefix that identifies lines as statements. + /// \param delimiter_ Delimiter to surround a variable instantiation. + templates_parser(const text::templates_def& templates_, + const std::string& prefix_, + const std::string& delimiter_) : + _templates(templates_), + _prefix(prefix_), + _delimiter(delimiter_), + _if_level(0), + _exit_if_level(0), + _loop_level(0), + _exit_loop_level(0) + { + } + + /// Applies the templates to a given input. + /// + /// \param input The stream to which to apply the templates. + /// \param output The stream into which to write the results. + /// + /// \throw text::syntax_error If the input is not valid. Note that the + /// is not guaranteed to be unmodified on exit if an error is + /// encountered. + void + instantiate(std::istream& input, std::ostream& output) + { + std::string line; + while (std::getline(input, line).good()) { + if (!_skip.empty() && _skip.top()) + handle_skip(line); + else + handle_normal(evaluate(line), input, output); + } + } +}; + + +} // anonymous namespace + + +/// Constructs an empty templates definition. +text::templates_def::templates_def(void) +{ +} + + +/// Sets a string variable in the templates. +/// +/// If the variable already exists, its value is replaced. This behavior is +/// required to implement iterators, but client code should really not be +/// redefining variables. +/// +/// \pre The variable must not already exist as a vector. +/// +/// \param name The name of the variable to set. +/// \param value The value to set the given variable to. +void +text::templates_def::add_variable(const std::string& name, + const std::string& value) +{ + PRE(_vectors.find(name) == _vectors.end()); + _variables[name] = value; +} + + +/// Unsets a string variable from the templates. +/// +/// Client code has no reason to use this. This is only required to implement +/// proper scoping of loop iterators. +/// +/// \pre The variable must exist. +/// +/// \param name The name of the variable to remove from the templates. +void +text::templates_def::remove_variable(const std::string& name) +{ + PRE(_variables.find(name) != _variables.end()); + _variables.erase(_variables.find(name)); +} + + +/// Creates a new vector in the templates. +/// +/// If the vector already exists, it is cleared. Client code should really not +/// be redefining variables. +/// +/// \pre The vector must not already exist as a variable. +/// +/// \param name The name of the vector to set. +void +text::templates_def::add_vector(const std::string& name) +{ + PRE(_variables.find(name) == _variables.end()); + _vectors[name] = strings_vector(); +} + + +/// Adds a value to an existing vector in the templates. +/// +/// \pre name The vector must exist. +/// +/// \param name The name of the vector to append the value to. +/// \param value The textual value to append to the vector. +void +text::templates_def::add_to_vector(const std::string& name, + const std::string& value) +{ + PRE(_variables.find(name) == _variables.end()); + PRE(_vectors.find(name) != _vectors.end()); + _vectors[name].push_back(value); +} + + +/// Checks whether a given identifier exists as a variable or a vector. +/// +/// This is used to implement the evaluation of conditions in if clauses. +/// +/// \param name The name of the variable or vector. +/// +/// \return True if the given name exists as a variable or a vector; false +/// otherwise. +bool +text::templates_def::exists(const std::string& name) const +{ + return (_variables.find(name) != _variables.end() || + _vectors.find(name) != _vectors.end()); +} + + +/// Gets the value of a variable. +/// +/// \param name The name of the variable. +/// +/// \return The value of the requested variable. +/// +/// \throw text::syntax_error If the variable does not exist. +const std::string& +text::templates_def::get_variable(const std::string& name) const +{ + const variables_map::const_iterator iter = _variables.find(name); + if (iter == _variables.end()) + throw text::syntax_error(F("Unknown variable '%s'") % name); + return (*iter).second; +} + + +/// Gets a vector. +/// +/// \param name The name of the vector. +/// +/// \return A reference to the requested vector. +/// +/// \throw text::syntax_error If the vector does not exist. +const text::templates_def::strings_vector& +text::templates_def::get_vector(const std::string& name) const +{ + const vectors_map::const_iterator iter = _vectors.find(name); + if (iter == _vectors.end()) + throw text::syntax_error(F("Unknown vector '%s'") % name); + return (*iter).second; +} + + +/// Indexes a vector and gets the value. +/// +/// \param name The name of the vector to index. +/// \param index_name The name of a variable representing the index to use. +/// This must be convertible to a natural. +/// +/// \return The value of the vector at the given index. +/// +/// \throw text::syntax_error If the vector does not existor if the index is out +/// of range. +const std::string& +text::templates_def::get_vector(const std::string& name, + const std::string& index_name) const +{ + const strings_vector& vector = get_vector(name); + const std::string& index_str = get_variable(index_name); + + std::size_t index; + try { + index = text::to_type< std::size_t >(index_str); + } catch (const text::syntax_error& e) { + throw text::syntax_error(F("Index '%s' not an integer, value '%s'") % + index_name % index_str); + } + if (index >= vector.size()) + throw text::syntax_error(F("Index '%s' out of range at position '%s'") % + index_name % index); + + return vector[index]; +} + + +/// Evaluates a expression using these templates. +/// +/// An expression is a query on the current templates to fetch a particular +/// value. The value is always returned as a string, as this is how templates +/// are internally stored. +/// +/// \param expression The expression to evaluate. This should not include any +/// of the delimiters used in the user input, as otherwise the expression +/// will not be evaluated properly. +/// +/// \return The result of the expression evaluation as a string. +/// +/// \throw text::syntax_error If there is any problem while evaluating the +/// expression. +std::string +text::templates_def::evaluate(const std::string& expression) const +{ + const std::string::size_type paren_open = expression.find('('); + if (paren_open == std::string::npos) { + return get_variable(expression); + } else { + const std::string::size_type paren_close = expression.find( + ')', paren_open); + if (paren_close == std::string::npos) + throw text::syntax_error(F("Expected ')' in expression '%s')") % + expression); + if (paren_close != expression.length() - 1) + throw text::syntax_error(F("Unexpected text found after ')' in " + "expression '%s'") % expression); + + const std::string arg0 = expression.substr(0, paren_open); + const std::string arg1 = expression.substr( + paren_open + 1, paren_close - paren_open - 1); + if (arg0 == "defined") { + return exists(arg1) ? "true" : "false"; + } else if (arg0 == "length") { + return F("%s") % get_vector(arg1).size(); + } else { + return get_vector(arg0, arg1); + } + } +} + + +/// Applies a set of templates to an input stream. +/// +/// \param templates The templates to use. +/// \param input The input to process. +/// \param output The stream to which to write the processed text. +/// +/// \throw text::syntax_error If there is any problem processing the input. +void +text::instantiate(const templates_def& templates, + std::istream& input, std::ostream& output) +{ + templates_parser parser(templates, "%", "%%"); + parser.instantiate(input, output); +} + + +/// Applies a set of templates to an input file and writes an output file. +/// +/// \param templates The templates to use. +/// \param input_file The path to the input to process. +/// \param output_file The path to the file into which to write the output. +/// +/// \throw text::error If the input or output files cannot be opened. +/// \throw text::syntax_error If there is any problem processing the input. +void +text::instantiate(const templates_def& templates, + const fs::path& input_file, const fs::path& output_file) +{ + std::ifstream input(input_file.c_str()); + if (!input) + throw text::error(F("Failed to open %s for read") % input_file); + + std::ofstream output(output_file.c_str()); + if (!output) + throw text::error(F("Failed to open %s for write") % output_file); + + instantiate(templates, input, output); +} diff --git a/utils/text/templates.hpp b/utils/text/templates.hpp new file mode 100644 index 000000000000..ffbf28512d0d --- /dev/null +++ b/utils/text/templates.hpp @@ -0,0 +1,122 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/templates.hpp +/// Custom templating engine for text documents. +/// +/// This module provides a simple mechanism to generate text documents based on +/// templates. The templates are just text files that contain template +/// statements that instruct this processor to perform transformations on the +/// input. +/// +/// While this was originally written to handle HTML templates, it is actually +/// generic enough to handle any kind of text document, hence why it lives +/// within the utils::text library. +/// +/// An example of how the templates look like: +/// +/// %if names +/// List of names +/// ------------- +/// Amount of names: %%length(names)%% +/// Most preferred name: %%preferred_name%% +/// Full list: +/// %loop names iter +/// * %%last_names(iter)%%, %%names(iter)%% +/// %endloop +/// %endif names + +#if !defined(UTILS_TEXT_TEMPLATES_HPP) +#define UTILS_TEXT_TEMPLATES_HPP + +#include "utils/text/templates_fwd.hpp" + +#include <istream> +#include <map> +#include <ostream> +#include <string> +#include <vector> + +#include "utils/fs/path_fwd.hpp" + +namespace utils { +namespace text { + + +/// Definitions of the templates to apply to a file. +/// +/// This class provides the environment (e.g. the list of variables) that the +/// templating system has to use when generating the output files. This +/// definition is static in the sense that this is what the caller program +/// specifies. +class templates_def { + /// Mapping of variable names to their values. + typedef std::map< std::string, std::string > variables_map; + + /// Collection of global variables available to the templates. + variables_map _variables; + + /// Convenience name for a vector of strings. + typedef std::vector< std::string > strings_vector; + + /// Mapping of vector names to their contents. + /// + /// Ideally, these would be represented as part of the _variables, but we + /// would need a complex mechanism to identify whether a variable is a + /// string or a vector. + typedef std::map< std::string, strings_vector > vectors_map; + + /// Collection of vectors available to the templates. + vectors_map _vectors; + + const std::string& get_vector(const std::string&, const std::string&) const; + +public: + templates_def(void); + + void add_variable(const std::string&, const std::string&); + void remove_variable(const std::string&); + void add_vector(const std::string&); + void add_to_vector(const std::string&, const std::string&); + + bool exists(const std::string&) const; + const std::string& get_variable(const std::string&) const; + const strings_vector& get_vector(const std::string&) const; + + std::string evaluate(const std::string&) const; +}; + + +void instantiate(const templates_def&, std::istream&, std::ostream&); +void instantiate(const templates_def&, const fs::path&, const fs::path&); + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_TEMPLATES_HPP) diff --git a/utils/text/templates_fwd.hpp b/utils/text/templates_fwd.hpp new file mode 100644 index 000000000000..c806be0cf497 --- /dev/null +++ b/utils/text/templates_fwd.hpp @@ -0,0 +1,45 @@ +// Copyright 2015 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +/// \file utils/text/templates_fwd.hpp +/// Forward declarations for utils/text/templates.hpp + +#if !defined(UTILS_TEXT_TEMPLATES_FWD_HPP) +#define UTILS_TEXT_TEMPLATES_FWD_HPP + +namespace utils { +namespace text { + + +class templates_def; + + +} // namespace text +} // namespace utils + +#endif // !defined(UTILS_TEXT_TEMPLATES_FWD_HPP) diff --git a/utils/text/templates_test.cpp b/utils/text/templates_test.cpp new file mode 100644 index 000000000000..4524dc61a416 --- /dev/null +++ b/utils/text/templates_test.cpp @@ -0,0 +1,1001 @@ +// Copyright 2012 The Kyua Authors. +// All rights reserved. +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are +// met: +// +// * Redistributions of source code must retain the above copyright +// notice, this list of conditions and the following disclaimer. +// * Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// * Neither the name of Google Inc. nor the names of its contributors +// may be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +#include "utils/text/templates.hpp" + +#include <fstream> +#include <sstream> + +#include <atf-c++.hpp> + +#include "utils/fs/operations.hpp" +#include "utils/fs/path.hpp" +#include "utils/text/exceptions.hpp" + +namespace fs = utils::fs; +namespace text = utils::text; + + +namespace { + + +/// Applies a set of templates to an input string and validates the output. +/// +/// This fails the test case if exp_output does not match the document generated +/// by the application of the templates. +/// +/// \param templates The templates to apply. +/// \param input_str The input document to which to apply the templates. +/// \param exp_output The expected output document. +static void +do_test_ok(const text::templates_def& templates, const std::string& input_str, + const std::string& exp_output) +{ + std::istringstream input(input_str); + std::ostringstream output; + + text::instantiate(templates, input, output); + ATF_REQUIRE_EQ(exp_output, output.str()); +} + + +/// Applies a set of templates to an input string and checks for an error. +/// +/// This fails the test case if the exception raised by the template processing +/// does not match the expected message. +/// +/// \param templates The templates to apply. +/// \param input_str The input document to which to apply the templates. +/// \param exp_message The expected error message in the raised exception. +static void +do_test_fail(const text::templates_def& templates, const std::string& input_str, + const std::string& exp_message) +{ + std::istringstream input(input_str); + std::ostringstream output; + + ATF_REQUIRE_THROW_RE(text::syntax_error, exp_message, + text::instantiate(templates, input, output)); +} + + +} // anonymous namespace + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__add_variable__first); +ATF_TEST_CASE_BODY(templates_def__add_variable__first) +{ + text::templates_def templates; + templates.add_variable("the-name", "first-value"); + ATF_REQUIRE_EQ("first-value", templates.get_variable("the-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__add_variable__replace); +ATF_TEST_CASE_BODY(templates_def__add_variable__replace) +{ + text::templates_def templates; + templates.add_variable("the-name", "first-value"); + templates.add_variable("the-name", "second-value"); + ATF_REQUIRE_EQ("second-value", templates.get_variable("the-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__remove_variable); +ATF_TEST_CASE_BODY(templates_def__remove_variable) +{ + text::templates_def templates; + templates.add_variable("the-name", "the-value"); + templates.get_variable("the-name"); // Should not throw. + templates.remove_variable("the-name"); + ATF_REQUIRE_THROW(text::syntax_error, templates.get_variable("the-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__add_vector__first); +ATF_TEST_CASE_BODY(templates_def__add_vector__first) +{ + text::templates_def templates; + templates.add_vector("the-name"); + ATF_REQUIRE(templates.get_vector("the-name").empty()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__add_vector__replace); +ATF_TEST_CASE_BODY(templates_def__add_vector__replace) +{ + text::templates_def templates; + templates.add_vector("the-name"); + templates.add_to_vector("the-name", "foo"); + ATF_REQUIRE(!templates.get_vector("the-name").empty()); + templates.add_vector("the-name"); + ATF_REQUIRE(templates.get_vector("the-name").empty()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__add_to_vector); +ATF_TEST_CASE_BODY(templates_def__add_to_vector) +{ + text::templates_def templates; + templates.add_vector("the-name"); + ATF_REQUIRE_EQ(0, templates.get_vector("the-name").size()); + templates.add_to_vector("the-name", "first"); + ATF_REQUIRE_EQ(1, templates.get_vector("the-name").size()); + templates.add_to_vector("the-name", "second"); + ATF_REQUIRE_EQ(2, templates.get_vector("the-name").size()); + templates.add_to_vector("the-name", "third"); + ATF_REQUIRE_EQ(3, templates.get_vector("the-name").size()); + + std::vector< std::string > expected; + expected.push_back("first"); + expected.push_back("second"); + expected.push_back("third"); + ATF_REQUIRE(expected == templates.get_vector("the-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__exists__variable); +ATF_TEST_CASE_BODY(templates_def__exists__variable) +{ + text::templates_def templates; + ATF_REQUIRE(!templates.exists("some-name")); + templates.add_variable("some-name ", "foo"); + ATF_REQUIRE(!templates.exists("some-name")); + templates.add_variable("some-name", "foo"); + ATF_REQUIRE(templates.exists("some-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__exists__vector); +ATF_TEST_CASE_BODY(templates_def__exists__vector) +{ + text::templates_def templates; + ATF_REQUIRE(!templates.exists("some-name")); + templates.add_vector("some-name "); + ATF_REQUIRE(!templates.exists("some-name")); + templates.add_vector("some-name"); + ATF_REQUIRE(templates.exists("some-name")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__get_variable__ok); +ATF_TEST_CASE_BODY(templates_def__get_variable__ok) +{ + text::templates_def templates; + templates.add_variable("foo", ""); + templates.add_variable("bar", " baz "); + ATF_REQUIRE_EQ("", templates.get_variable("foo")); + ATF_REQUIRE_EQ(" baz ", templates.get_variable("bar")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__get_variable__unknown); +ATF_TEST_CASE_BODY(templates_def__get_variable__unknown) +{ + text::templates_def templates; + templates.add_variable("foo", ""); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown variable 'foo '", + templates.get_variable("foo ")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__get_vector__ok); +ATF_TEST_CASE_BODY(templates_def__get_vector__ok) +{ + text::templates_def templates; + templates.add_vector("foo"); + templates.add_vector("bar"); + templates.add_to_vector("bar", "baz"); + ATF_REQUIRE_EQ(0, templates.get_vector("foo").size()); + ATF_REQUIRE_EQ(1, templates.get_vector("bar").size()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__get_vector__unknown); +ATF_TEST_CASE_BODY(templates_def__get_vector__unknown) +{ + text::templates_def templates; + templates.add_vector("foo"); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown vector 'foo '", + templates.get_vector("foo ")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__variable__ok); +ATF_TEST_CASE_BODY(templates_def__evaluate__variable__ok) +{ + text::templates_def templates; + templates.add_variable("foo", ""); + templates.add_variable("bar", " baz "); + ATF_REQUIRE_EQ("", templates.evaluate("foo")); + ATF_REQUIRE_EQ(" baz ", templates.evaluate("bar")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__variable__unknown); +ATF_TEST_CASE_BODY(templates_def__evaluate__variable__unknown) +{ + text::templates_def templates; + templates.add_variable("foo", ""); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown variable 'foo1'", + templates.evaluate("foo1")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__vector__ok); +ATF_TEST_CASE_BODY(templates_def__evaluate__vector__ok) +{ + text::templates_def templates; + templates.add_vector("v"); + templates.add_to_vector("v", "foo"); + templates.add_to_vector("v", "bar"); + templates.add_to_vector("v", "baz"); + + templates.add_variable("index", "0"); + ATF_REQUIRE_EQ("foo", templates.evaluate("v(index)")); + templates.add_variable("index", "1"); + ATF_REQUIRE_EQ("bar", templates.evaluate("v(index)")); + templates.add_variable("index", "2"); + ATF_REQUIRE_EQ("baz", templates.evaluate("v(index)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__vector__unknown_vector); +ATF_TEST_CASE_BODY(templates_def__evaluate__vector__unknown_vector) +{ + text::templates_def templates; + templates.add_vector("v"); + templates.add_to_vector("v", "foo"); + templates.add_variable("index", "0"); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown vector 'fooz'", + templates.evaluate("fooz(index)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__vector__unknown_index); +ATF_TEST_CASE_BODY(templates_def__evaluate__vector__unknown_index) +{ + text::templates_def templates; + templates.add_vector("v"); + templates.add_to_vector("v", "foo"); + templates.add_variable("index", "0"); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown variable 'indexz'", + templates.evaluate("v(indexz)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__vector__out_of_range); +ATF_TEST_CASE_BODY(templates_def__evaluate__vector__out_of_range) +{ + text::templates_def templates; + templates.add_vector("v"); + templates.add_to_vector("v", "foo"); + templates.add_variable("index", "1"); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Index 'index' out of range " + "at position '1'", templates.evaluate("v(index)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__defined); +ATF_TEST_CASE_BODY(templates_def__evaluate__defined) +{ + text::templates_def templates; + templates.add_vector("the-variable"); + templates.add_vector("the-vector"); + ATF_REQUIRE_EQ("false", templates.evaluate("defined(the-variabl)")); + ATF_REQUIRE_EQ("false", templates.evaluate("defined(the-vecto)")); + ATF_REQUIRE_EQ("true", templates.evaluate("defined(the-variable)")); + ATF_REQUIRE_EQ("true", templates.evaluate("defined(the-vector)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__length__ok); +ATF_TEST_CASE_BODY(templates_def__evaluate__length__ok) +{ + text::templates_def templates; + templates.add_vector("v"); + ATF_REQUIRE_EQ("0", templates.evaluate("length(v)")); + templates.add_to_vector("v", "foo"); + ATF_REQUIRE_EQ("1", templates.evaluate("length(v)")); + templates.add_to_vector("v", "bar"); + ATF_REQUIRE_EQ("2", templates.evaluate("length(v)")); + templates.add_to_vector("v", "baz"); + ATF_REQUIRE_EQ("3", templates.evaluate("length(v)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__length__unknown_vector); +ATF_TEST_CASE_BODY(templates_def__evaluate__length__unknown_vector) +{ + text::templates_def templates; + templates.add_vector("foo1"); + ATF_REQUIRE_THROW_RE(text::syntax_error, "Unknown vector 'foo'", + templates.evaluate("length(foo)")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(templates_def__evaluate__parenthesis_error); +ATF_TEST_CASE_BODY(templates_def__evaluate__parenthesis_error) +{ + text::templates_def templates; + ATF_REQUIRE_THROW_RE(text::syntax_error, + "Expected '\\)' in.*'foo\\(abc'", + templates.evaluate("foo(abc")); + ATF_REQUIRE_THROW_RE(text::syntax_error, + "Unexpected text.*'\\)' in.*'a\\(b\\)c'", + templates.evaluate("a(b)c")); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__empty_input); +ATF_TEST_CASE_BODY(instantiate__empty_input) +{ + const text::templates_def templates; + do_test_ok(templates, "", ""); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__value__ok); +ATF_TEST_CASE_BODY(instantiate__value__ok) +{ + const std::string input = + "first line\n" + "%%testvar1%%\n" + "third line\n" + "%%testvar2%% %%testvar3%%%%testvar4%%\n" + "fifth line\n"; + + const std::string exp_output = + "first line\n" + "second line\n" + "third line\n" + "fourth line.\n" + "fifth line\n"; + + text::templates_def templates; + templates.add_variable("testvar1", "second line"); + templates.add_variable("testvar2", "fourth"); + templates.add_variable("testvar3", "line"); + templates.add_variable("testvar4", "."); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__value__unknown_variable); +ATF_TEST_CASE_BODY(instantiate__value__unknown_variable) +{ + const std::string input = + "%%testvar1%%\n"; + + text::templates_def templates; + templates.add_variable("testvar2", "fourth line"); + + do_test_fail(templates, input, "Unknown variable 'testvar1'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_length__ok); +ATF_TEST_CASE_BODY(instantiate__vector_length__ok) +{ + const std::string input = + "%%length(testvector1)%%\n" + "%%length(testvector2)%% - %%length(testvector3)%%\n"; + + const std::string exp_output = + "4\n" + "0 - 1\n"; + + text::templates_def templates; + templates.add_vector("testvector1"); + templates.add_to_vector("testvector1", "000"); + templates.add_to_vector("testvector1", "111"); + templates.add_to_vector("testvector1", "543"); + templates.add_to_vector("testvector1", "999"); + templates.add_vector("testvector2"); + templates.add_vector("testvector3"); + templates.add_to_vector("testvector3", "123"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_length__unknown_vector); +ATF_TEST_CASE_BODY(instantiate__vector_length__unknown_vector) +{ + const std::string input = + "%%length(testvector)%%\n"; + + text::templates_def templates; + templates.add_vector("testvector2"); + + do_test_fail(templates, input, "Unknown vector 'testvector'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_value__ok); +ATF_TEST_CASE_BODY(instantiate__vector_value__ok) +{ + const std::string input = + "first line\n" + "%%testvector1(i)%%\n" + "third line\n" + "%%testvector2(j)%%\n" + "fifth line\n"; + + const std::string exp_output = + "first line\n" + "543\n" + "third line\n" + "123\n" + "fifth line\n"; + + text::templates_def templates; + templates.add_variable("i", "2"); + templates.add_variable("j", "0"); + templates.add_vector("testvector1"); + templates.add_to_vector("testvector1", "000"); + templates.add_to_vector("testvector1", "111"); + templates.add_to_vector("testvector1", "543"); + templates.add_to_vector("testvector1", "999"); + templates.add_vector("testvector2"); + templates.add_to_vector("testvector2", "123"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_value__unknown_vector); +ATF_TEST_CASE_BODY(instantiate__vector_value__unknown_vector) +{ + const std::string input = + "%%testvector(j)%%\n"; + + text::templates_def templates; + templates.add_vector("testvector2"); + + do_test_fail(templates, input, "Unknown vector 'testvector'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_value__out_of_range__empty); +ATF_TEST_CASE_BODY(instantiate__vector_value__out_of_range__empty) +{ + const std::string input = + "%%testvector(j)%%\n"; + + text::templates_def templates; + templates.add_vector("testvector"); + templates.add_variable("j", "0"); + + do_test_fail(templates, input, "Index 'j' out of range at position '0'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__vector_value__out_of_range__not_empty); +ATF_TEST_CASE_BODY(instantiate__vector_value__out_of_range__not_empty) +{ + const std::string input = + "%%testvector(j)%%\n"; + + text::templates_def templates; + templates.add_vector("testvector"); + templates.add_to_vector("testvector", "a"); + templates.add_to_vector("testvector", "b"); + templates.add_variable("j", "2"); + + do_test_fail(templates, input, "Index 'j' out of range at position '2'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__if__one_level__taken); +ATF_TEST_CASE_BODY(instantiate__if__one_level__taken) +{ + const std::string input = + "first line\n" + "%if defined(some_var)\n" + "hello from within the variable conditional\n" + "%endif\n" + "%if defined(some_vector)\n" + "hello from within the vector conditional\n" + "%else\n" + "bye from within the vector conditional\n" + "%endif\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "hello from within the variable conditional\n" + "hello from within the vector conditional\n" + "some more\n"; + + text::templates_def templates; + templates.add_variable("some_var", "zzz"); + templates.add_vector("some_vector"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__if__one_level__not_taken); +ATF_TEST_CASE_BODY(instantiate__if__one_level__not_taken) +{ + const std::string input = + "first line\n" + "%if defined(some_var)\n" + "hello from within the variable conditional\n" + "%endif\n" + "%if defined(some_vector)\n" + "hello from within the vector conditional\n" + "%else\n" + "bye from within the vector conditional\n" + "%endif\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "bye from within the vector conditional\n" + "some more\n"; + + text::templates_def templates; + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__if__multiple_levels__taken); +ATF_TEST_CASE_BODY(instantiate__if__multiple_levels__taken) +{ + const std::string input = + "first line\n" + "%if defined(var1)\n" + "first before\n" + "%if length(var2)\n" + "second before\n" + "%if defined(var3)\n" + "third before\n" + "hello from within the conditional\n" + "third after\n" + "%endif\n" + "second after\n" + "%else\n" + "second after not shown\n" + "%endif\n" + "first after\n" + "%endif\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "first before\n" + "second before\n" + "third before\n" + "hello from within the conditional\n" + "third after\n" + "second after\n" + "first after\n" + "some more\n"; + + text::templates_def templates; + templates.add_variable("var1", "false"); + templates.add_vector("var2"); + templates.add_to_vector("var2", "not-empty"); + templates.add_variable("var3", "foobar"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__if__multiple_levels__not_taken); +ATF_TEST_CASE_BODY(instantiate__if__multiple_levels__not_taken) +{ + const std::string input = + "first line\n" + "%if defined(var1)\n" + "first before\n" + "%if length(var2)\n" + "second before\n" + "%if defined(var3)\n" + "third before\n" + "hello from within the conditional\n" + "third after\n" + "%else\n" + "will not be shown either\n" + "%endif\n" + "second after\n" + "%else\n" + "second after shown\n" + "%endif\n" + "first after\n" + "%endif\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "first before\n" + "second after shown\n" + "first after\n" + "some more\n"; + + text::templates_def templates; + templates.add_variable("var1", "false"); + templates.add_vector("var2"); + templates.add_vector("var3"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__no_iterations); +ATF_TEST_CASE_BODY(instantiate__loop__no_iterations) +{ + const std::string input = + "first line\n" + "%loop table1 i\n" + "hello\n" + "value in vector: %%table1(i)%%\n" + "%if defined(var1)\n" "some other text\n" "%endif\n" + "%endloop\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "some more\n"; + + text::templates_def templates; + templates.add_variable("var1", "defined"); + templates.add_vector("table1"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__multiple_iterations); +ATF_TEST_CASE_BODY(instantiate__loop__multiple_iterations) +{ + const std::string input = + "first line\n" + "%loop table1 i\n" + "hello %%table1(i)%% %%table2(i)%%\n" + "%endloop\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "hello foo1 foo2\n" + "hello bar1 bar2\n" + "some more\n"; + + text::templates_def templates; + templates.add_vector("table1"); + templates.add_to_vector("table1", "foo1"); + templates.add_to_vector("table1", "bar1"); + templates.add_vector("table2"); + templates.add_to_vector("table2", "foo2"); + templates.add_to_vector("table2", "bar2"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__nested__no_iterations); +ATF_TEST_CASE_BODY(instantiate__loop__nested__no_iterations) +{ + const std::string input = + "first line\n" + "%loop table1 i\n" + "before: %%table1(i)%%\n" + "%loop table2 j\n" + "before: %%table2(j)%%\n" + "%loop table3 k\n" + "%%table3(k)%%\n" + "%endloop\n" + "after: %%table2(i)%%\n" + "%endloop\n" + "after: %%table1(i)%%\n" + "%endloop\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "before: a\n" + "after: a\n" + "before: b\n" + "after: b\n" + "some more\n"; + + text::templates_def templates; + templates.add_vector("table1"); + templates.add_to_vector("table1", "a"); + templates.add_to_vector("table1", "b"); + templates.add_vector("table2"); + templates.add_vector("table3"); + templates.add_to_vector("table3", "1"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__nested__multiple_iterations); +ATF_TEST_CASE_BODY(instantiate__loop__nested__multiple_iterations) +{ + const std::string input = + "first line\n" + "%loop table1 i\n" + "%loop table2 j\n" + "%%table1(i)%% %%table2(j)%%\n" + "%endloop\n" + "%endloop\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "a 1\n" + "a 2\n" + "a 3\n" + "b 1\n" + "b 2\n" + "b 3\n" + "some more\n"; + + text::templates_def templates; + templates.add_vector("table1"); + templates.add_to_vector("table1", "a"); + templates.add_to_vector("table1", "b"); + templates.add_vector("table2"); + templates.add_to_vector("table2", "1"); + templates.add_to_vector("table2", "2"); + templates.add_to_vector("table2", "3"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__sequential); +ATF_TEST_CASE_BODY(instantiate__loop__sequential) +{ + const std::string input = + "first line\n" + "%loop table1 iter\n" + "1: %%table1(iter)%%\n" + "%endloop\n" + "divider\n" + "%loop table2 iter\n" + "2: %%table2(iter)%%\n" + "%endloop\n" + "divider\n" + "%loop table3 iter\n" + "3: %%table3(iter)%%\n" + "%endloop\n" + "divider\n" + "%loop table4 iter\n" + "4: %%table4(iter)%%\n" + "%endloop\n" + "some more\n"; + + const std::string exp_output = + "first line\n" + "1: a\n" + "1: b\n" + "divider\n" + "divider\n" + "divider\n" + "4: 1\n" + "4: 2\n" + "4: 3\n" + "some more\n"; + + text::templates_def templates; + templates.add_vector("table1"); + templates.add_to_vector("table1", "a"); + templates.add_to_vector("table1", "b"); + templates.add_vector("table2"); + templates.add_vector("table3"); + templates.add_vector("table4"); + templates.add_to_vector("table4", "1"); + templates.add_to_vector("table4", "2"); + templates.add_to_vector("table4", "3"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__loop__scoping); +ATF_TEST_CASE_BODY(instantiate__loop__scoping) +{ + const std::string input = + "%loop table1 i\n" + "%if defined(i)\n" "i defined inside scope 1\n" "%endif\n" + "%loop table2 j\n" + "%if defined(i)\n" "i defined inside scope 2\n" "%endif\n" + "%if defined(j)\n" "j defined inside scope 2\n" "%endif\n" + "%endloop\n" + "%if defined(j)\n" "j defined inside scope 1\n" "%endif\n" + "%endloop\n" + "%if defined(i)\n" "i defined outside\n" "%endif\n" + "%if defined(j)\n" "j defined outside\n" "%endif\n"; + + const std::string exp_output = + "i defined inside scope 1\n" + "i defined inside scope 2\n" + "j defined inside scope 2\n" + "i defined inside scope 1\n" + "i defined inside scope 2\n" + "j defined inside scope 2\n"; + + text::templates_def templates; + templates.add_vector("table1"); + templates.add_to_vector("table1", "first"); + templates.add_to_vector("table1", "second"); + templates.add_vector("table2"); + templates.add_to_vector("table2", "first"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__mismatched_delimiters); +ATF_TEST_CASE_BODY(instantiate__mismatched_delimiters) +{ + const std::string input = + "this is some %% text\n" + "and this is %%var%% text%%\n"; + + const std::string exp_output = + "this is some %% text\n" + "and this is some more text%%\n"; + + text::templates_def templates; + templates.add_variable("var", "some more"); + + do_test_ok(templates, input, exp_output); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__empty_statement); +ATF_TEST_CASE_BODY(instantiate__empty_statement) +{ + do_test_fail(text::templates_def(), "%\n", "Empty statement"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__unknown_statement); +ATF_TEST_CASE_BODY(instantiate__unknown_statement) +{ + do_test_fail(text::templates_def(), "%if2\n", "Unknown statement 'if2'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__invalid_narguments); +ATF_TEST_CASE_BODY(instantiate__invalid_narguments) +{ + do_test_fail(text::templates_def(), "%if a b\n", + "Invalid number of arguments for statement 'if'"); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__files__ok); +ATF_TEST_CASE_BODY(instantiate__files__ok) +{ + text::templates_def templates; + templates.add_variable("string", "Hello, world!"); + + atf::utils::create_file("input.txt", "The string is: %%string%%\n"); + + text::instantiate(templates, fs::path("input.txt"), fs::path("output.txt")); + + std::ifstream output("output.txt"); + std::string line; + ATF_REQUIRE(std::getline(output, line).good()); + ATF_REQUIRE_EQ(line, "The string is: Hello, world!"); + ATF_REQUIRE(std::getline(output, line).eof()); +} + + +ATF_TEST_CASE_WITHOUT_HEAD(instantiate__files__input_error); +ATF_TEST_CASE_BODY(instantiate__files__input_error) +{ + text::templates_def templates; + ATF_REQUIRE_THROW_RE(text::error, "Failed to open input.txt for read", + text::instantiate(templates, fs::path("input.txt"), + fs::path("output.txt"))); +} + + +ATF_TEST_CASE(instantiate__files__output_error); +ATF_TEST_CASE_HEAD(instantiate__files__output_error) +{ + set_md_var("require.user", "unprivileged"); +} +ATF_TEST_CASE_BODY(instantiate__files__output_error) +{ + text::templates_def templates; + + atf::utils::create_file("input.txt", ""); + + fs::mkdir(fs::path("dir"), 0444); + + ATF_REQUIRE_THROW_RE(text::error, "Failed to open dir/output.txt for write", + text::instantiate(templates, fs::path("input.txt"), + fs::path("dir/output.txt"))); +} + + +ATF_INIT_TEST_CASES(tcs) +{ + ATF_ADD_TEST_CASE(tcs, templates_def__add_variable__first); + ATF_ADD_TEST_CASE(tcs, templates_def__add_variable__replace); + ATF_ADD_TEST_CASE(tcs, templates_def__remove_variable); + ATF_ADD_TEST_CASE(tcs, templates_def__add_vector__first); + ATF_ADD_TEST_CASE(tcs, templates_def__add_vector__replace); + ATF_ADD_TEST_CASE(tcs, templates_def__add_to_vector); + ATF_ADD_TEST_CASE(tcs, templates_def__exists__variable); + ATF_ADD_TEST_CASE(tcs, templates_def__exists__vector); + ATF_ADD_TEST_CASE(tcs, templates_def__get_variable__ok); + ATF_ADD_TEST_CASE(tcs, templates_def__get_variable__unknown); + ATF_ADD_TEST_CASE(tcs, templates_def__get_vector__ok); + ATF_ADD_TEST_CASE(tcs, templates_def__get_vector__unknown); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__variable__ok); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__variable__unknown); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__vector__ok); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__vector__unknown_vector); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__vector__unknown_index); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__vector__out_of_range); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__defined); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__length__ok); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__length__unknown_vector); + ATF_ADD_TEST_CASE(tcs, templates_def__evaluate__parenthesis_error); + + ATF_ADD_TEST_CASE(tcs, instantiate__empty_input); + ATF_ADD_TEST_CASE(tcs, instantiate__value__ok); + ATF_ADD_TEST_CASE(tcs, instantiate__value__unknown_variable); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_length__ok); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_length__unknown_vector); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_value__ok); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_value__unknown_vector); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_value__out_of_range__empty); + ATF_ADD_TEST_CASE(tcs, instantiate__vector_value__out_of_range__not_empty); + ATF_ADD_TEST_CASE(tcs, instantiate__if__one_level__taken); + ATF_ADD_TEST_CASE(tcs, instantiate__if__one_level__not_taken); + ATF_ADD_TEST_CASE(tcs, instantiate__if__multiple_levels__taken); + ATF_ADD_TEST_CASE(tcs, instantiate__if__multiple_levels__not_taken); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__no_iterations); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__multiple_iterations); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__nested__no_iterations); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__nested__multiple_iterations); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__sequential); + ATF_ADD_TEST_CASE(tcs, instantiate__loop__scoping); + ATF_ADD_TEST_CASE(tcs, instantiate__mismatched_delimiters); + ATF_ADD_TEST_CASE(tcs, instantiate__empty_statement); + ATF_ADD_TEST_CASE(tcs, instantiate__unknown_statement); + ATF_ADD_TEST_CASE(tcs, instantiate__invalid_narguments); + + ATF_ADD_TEST_CASE(tcs, instantiate__files__ok); + ATF_ADD_TEST_CASE(tcs, instantiate__files__input_error); + ATF_ADD_TEST_CASE(tcs, instantiate__files__output_error); +} |