src/testutils/refdata_xml.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2015,2016,2017,2019, by the GROMACS development team, led by
   5  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   6  * and including many others, as listed in the AUTHORS file in the
   7  * top-level source directory and at http://www.gromacs.org.
   8  *
   9  * GROMACS is free software; you can redistribute it and/or
  10  * modify it under the terms of the GNU Lesser General Public License
  11  * as published by the Free Software Foundation; either version 2.1
  12  * of the License, or (at your option) any later version.
  13  *
  14  * GROMACS is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  17  * Lesser General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU Lesser General Public
  20  * License along with GROMACS; if not, see
  21  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  22  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  23  *
  24  * If you want to redistribute modifications to GROMACS, please
  25  * consider that scientific software is very special. Version
  26  * control is crucial - bugs must be traceable. We will be happy to
  27  * consider code for inclusion in the official distribution, but
  28  * derived work must not be called official GROMACS. Details are found
  29  * in the README & COPYING files - if they are missing, get the
  30  * official version at http://www.gromacs.org.
  31  *
  32  * To help us fund GROMACS development, we humbly ask that you cite
  33  * the research papers on the package. Check out http://www.gromacs.org.
  34  */
  35 /*! \internal \file
  36  * \brief
  37  * Implements reference data XML persistence.
  38  *
  39  * \author Teemu Murtola <teemu.murtola@gmail.com>
  40  * \author Mark Abraham <mark.j.abraham@gmail.com>
  41  * \ingroup module_testutils
  42  */
  43 #include "gmxpre.h"
  44
  45 #include "refdata_xml.h"
  46
  47 #include <tinyxml2.h>
  48
  49 #include "gromacs/utility/exceptions.h"
  50
  51 #include "testutils/refdata_impl.h"
  52 #include "testutils/testexceptions.h"
  53
  54 namespace gmx
  55 {
  56 namespace test
  57 {
  58
  59 namespace
  60 {
  61
  62 //! XML version declaration used when writing the reference data.
  63 const char* const c_VersionDeclarationString = "xml version=\"1.0\"";
  64 //! XML stylesheet declaration used for writing the reference data.
  65 const char* const c_StyleSheetDeclarationString =
  66         "xml-stylesheet type=\"text/xsl\" href=\"referencedata.xsl\"";
  67 //! Name of the root element in reference data XML files.
  68 const char* const c_RootNodeName = "ReferenceData";
  69 //! Name of the XML attribute used to store identifying strings for reference data elements.
  70 const char* const c_IdAttrName = "Name";
  71
  72 } // namespace
  73
  74 /********************************************************************
  75  * XML reading
  76  */
  77
  78 namespace
  79 {
  80
  81 //! Convenience typedef
  82 typedef tinyxml2::XMLDocument* XMLDocumentPtr;
  83 //! Convenience typedef
  84 typedef tinyxml2::XMLNode* XMLNodePtr;
  85 //! Convenience typedef
  86 typedef tinyxml2::XMLElement* XMLElementPtr;
  87 //! Convenience typedef
  88 typedef tinyxml2::XMLText* XMLTextPtr;
  89
  90 //! \name Helper functions for XML reading
  91 //! \{
  92
  93 void readEntry(XMLNodePtr node, ReferenceDataEntry* entry);
  94
  95 XMLNodePtr getCDataChildNode(XMLNodePtr node)
  96 {
  97     XMLNodePtr cdata = node->FirstChild();
  98     while (cdata != nullptr && cdata->ToText() != nullptr && !cdata->ToText()->CData())
  99     {
 100         cdata = cdata->NextSibling();
 101     }
 102     return cdata;
 103 }
 104
 105 bool hasCDataContent(XMLNodePtr node)
 106 {
 107     return getCDataChildNode(node) != nullptr;
 108 }
 109
 110 //! Return a node convertible to text, either \c childNode or its first such sibling.
 111 XMLNodePtr getNextTextChildNode(XMLNodePtr childNode)
 112 {
 113     // Note that when reading, we don't have to care if it is in a
 114     // CDATA section, or not.
 115     while (childNode != nullptr)
 116     {
 117         if (childNode->ToText() != nullptr)
 118         {
 119             break;
 120         }
 121         childNode = childNode->NextSibling();
 122     }
 123     return childNode;
 124 }
 125
 126 //! Return the concatenation of all the text children of \c node, including multiple CDATA children.
 127 std::string getValueFromLeafElement(XMLNodePtr node)
 128 {
 129     std::string value;
 130
 131     XMLNodePtr childNode = getNextTextChildNode(node->FirstChild());
 132     while (childNode != nullptr)
 133     {
 134         value += std::string(childNode->Value());
 135
 136         childNode = getNextTextChildNode(childNode->NextSibling());
 137     }
 138
 139     if (hasCDataContent(node))
 140     {
 141         // Prepare to strip the convenience newline added in
 142         // createElementContents, when writing CDATA content for
 143         // StringBlock data.
 144         if (value.empty() || value[0] != '\n')
 145         {
 146             GMX_THROW(TestException("Invalid string block in reference data"));
 147         }
 148         value.erase(0, 1);
 149     }
 150
 151     return value;
 152 }
 153
 154 //! Make a new entry from \c element.
 155 ReferenceDataEntry::EntryPointer createEntry(XMLElementPtr element)
 156 {
 157     const char*                      id = element->Attribute(c_IdAttrName);
 158     ReferenceDataEntry::EntryPointer entry(new ReferenceDataEntry(element->Value(), id));
 159     return entry;
 160 }
 161
 162 //! Read the child entries of \c parentElement and transfer the contents to \c entry
 163 void readChildEntries(XMLNodePtr parentElement, ReferenceDataEntry* entry)
 164 {
 165     XMLElementPtr childElement = parentElement->FirstChildElement();
 166     while (childElement != nullptr)
 167     {
 168         ReferenceDataEntry::EntryPointer child(createEntry(childElement));
 169         readEntry(childElement, child.get());
 170         entry->addChild(move(child));
 171         childElement = childElement->NextSiblingElement();
 172     }
 173 }
 174
 175 //! Return whether \c node has child XML elements (rather than text content).
 176 bool isCompoundElement(XMLNodePtr node)
 177 {
 178     return node->FirstChildElement() != nullptr;
 179 }
 180
 181 //! Read \c element and transfer the contents to \c entry
 182 void readEntry(XMLNodePtr element, ReferenceDataEntry* entry)
 183 {
 184     if (isCompoundElement(element))
 185     {
 186         readChildEntries(element, entry);
 187     }
 188     else if (hasCDataContent(element))
 189     {
 190         entry->setTextBlockValue(getValueFromLeafElement(element));
 191     }
 192     else
 193     {
 194         entry->setValue(getValueFromLeafElement(element));
 195     }
 196 }
 197
 198 //! \}
 199
 200 } // namespace
 201
 202 //! \cond internal
 203 ReferenceDataEntry::EntryPointer readReferenceDataFile(const std::string& path)
 204 {
 205     tinyxml2::XMLDocument document;
 206     document.LoadFile(path.c_str());
 207     if (document.Error())
 208     {
 209         const char* errorStr1 = document.GetErrorStr1();
 210         const char* errorStr2 = document.GetErrorStr2();
 211         std::string errorString("Error was ");
 212         if (errorStr1)
 213         {
 214             errorString += errorStr1;
 215         }
 216         if (errorStr2)
 217         {
 218             errorString += errorStr2;
 219         }
 220         if (!errorStr1 && !errorStr2)
 221         {
 222             errorString += "not specified.";
 223         }
 224         GMX_THROW(TestException("Reference data not parsed successfully: " + path + "\n."
 225                                 + errorString + "\n"));
 226     }
 227     XMLElementPtr rootNode = document.RootElement();
 228     if (rootNode == nullptr)
 229     {
 230         GMX_THROW(TestException("Reference data is empty: " + path));
 231     }
 232     if (std::strcmp(rootNode->Value(), c_RootNodeName) != 0)
 233     {
 234         GMX_THROW(TestException("Invalid root node type in " + path));
 235     }
 236
 237     ReferenceDataEntry::EntryPointer rootEntry(ReferenceDataEntry::createRoot());
 238     readEntry(rootNode, rootEntry.get());
 239     return rootEntry;
 240 }
 241 //! \endcond
 242
 243 /********************************************************************
 244  * XML writing
 245  */
 246
 247 namespace
 248 {
 249
 250 //! \name Helper functions for XML writing
 251 //! \{
 252
 253 void createElementAndContents(XMLElementPtr parentElement, const ReferenceDataEntry& entry);
 254
 255 void setIdAttribute(XMLElementPtr element, const std::string& id)
 256 {
 257     if (!id.empty())
 258     {
 259         element->SetAttribute(c_IdAttrName, id.c_str()); // If this fails, it throws std::bad_alloc
 260     }
 261 }
 262
 263 XMLElementPtr createElement(XMLElementPtr parentElement, const ReferenceDataEntry& entry)
 264 {
 265     XMLElementPtr element = parentElement->GetDocument()->NewElement(entry.type().c_str());
 266     parentElement->InsertEndChild(element);
 267     setIdAttribute(element, entry.id()); // If this fails, it throws std::bad_alloc
 268     return element;
 269 }
 270
 271 void createChildElements(XMLElementPtr parentElement, const ReferenceDataEntry& entry)
 272 {
 273     const ReferenceDataEntry::ChildList& children(entry.children());
 274     ReferenceDataEntry::ChildIterator    child;
 275     for (child = children.begin(); child != children.end(); ++child)
 276     {
 277         createElementAndContents(parentElement, **child);
 278     }
 279 }
 280
 281 /*! \brief Handle \c input intended to be written as CDATA
 282  *
 283  * This method searches for any ']]>' sequences embedded in \c input,
 284  * because this must always end a CDATA field. If any are found, it
 285  * breaks the string so that instead multiple CDATA fields will be
 286  * written with that token sequence split across the fields. Note that
 287  * tinyxml2 does not handle such things itself.
 288  *
 289  * This is an edge case that is unimportant for GROMACS refdata, but
 290  * it is preferable to know that the infrastructure won't break.
 291  */
 292 std::vector<std::string> breakUpAnyCdataEndTags(const std::string& input)
 293 {
 294     std::vector<std::string> strings;
 295     std::size_t              startPos = 0;
 296     std::size_t              endPos;
 297
 298     do
 299     {
 300         endPos = input.find("]]>", startPos);
 301         if (endPos != std::string::npos)
 302         {
 303             // We found an embedded CDATA end tag, so arrange to split it into multiple CDATA blocks
 304             endPos++;
 305         }
 306         strings.push_back(input.substr(startPos, endPos));
 307         startPos = endPos;
 308     } while (endPos != std::string::npos);
 309
 310     return strings;
 311 }
 312
 313 void createElementContents(XMLElementPtr element, const ReferenceDataEntry& entry)
 314 {
 315     // TODO: Figure out if \r and \r\n can be handled without them
 316     // changing to \n in the roundtrip.
 317     if (entry.isCompound())
 318     {
 319         createChildElements(element, entry);
 320     }
 321     else if (entry.isTextBlock())
 322     {
 323         // An extra newline is written in the beginning to make lines align
 324         // in the output xml (otherwise, the first line would be off by the length
 325         // of the starting CDATA tag).
 326         const std::string        adjustedValue = "\n" + entry.value();
 327         std::vector<std::string> cdataStrings  = breakUpAnyCdataEndTags(adjustedValue);
 328         for (auto const& s : cdataStrings)
 329         {
 330             XMLTextPtr textNode = element->GetDocument()->NewText(s.c_str());
 331             textNode->SetCData(true);
 332             element->InsertEndChild(textNode);
 333         }
 334     }
 335     else
 336     {
 337         XMLTextPtr textNode = element->GetDocument()->NewText(entry.value().c_str());
 338         element->InsertEndChild(textNode);
 339     }
 340 }
 341
 342 void createElementAndContents(XMLElementPtr parentElement, const ReferenceDataEntry& entry)
 343 {
 344     XMLElementPtr element = createElement(parentElement, entry);
 345     createElementContents(element, entry);
 346 }
 347
 348 XMLElementPtr createRootElement(XMLDocumentPtr document)
 349 {
 350     XMLElementPtr rootElement = document->NewElement(c_RootNodeName);
 351     document->InsertEndChild(rootElement);
 352     return rootElement;
 353 }
 354
 355 //! \}
 356
 357 } // namespace
 358
 359 //! \cond internal
 360 void writeReferenceDataFile(const std::string& path, const ReferenceDataEntry& rootEntry)
 361 {
 362     // TODO: Error checking
 363     tinyxml2::XMLDocument document;
 364
 365     tinyxml2::XMLDeclaration* declaration = document.NewDeclaration(c_VersionDeclarationString);
 366     document.InsertEndChild(declaration);
 367
 368     declaration = document.NewDeclaration(c_StyleSheetDeclarationString);
 369     document.InsertEndChild(declaration);
 370
 371     XMLElementPtr rootElement = createRootElement(&document);
 372     createChildElements(rootElement, rootEntry);
 373
 374     if (document.SaveFile(path.c_str()) != tinyxml2::XML_NO_ERROR)
 375     {
 376         GMX_THROW(TestException("Reference data saving failed in " + path));
 377     }
 378 }
 379 //! \endcond
 380
 381 } // namespace test
 382 } // namespace gmx