src/testutils/refdata-xml.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 2015,2016,2017, 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 = "xml-stylesheet type=\"text/xsl\" href=\"referencedata.xsl\"";
  66 //! Name of the root element in reference data XML files.
  67 const char *const c_RootNodeName = "ReferenceData";
  68 //! Name of the XML attribute used to store identifying strings for reference data elements.
  69 const char *const c_IdAttrName = "Name";
  70
  71 }       // namespace
  72
  73 /********************************************************************
  74  * XML reading
  75  */
  76
  77 namespace
  78 {
  79
  80 //! Convenience typedef
  81 typedef tinyxml2::XMLDocument *XMLDocumentPtr;
  82 //! Convenience typedef
  83 typedef tinyxml2::XMLNode *XMLNodePtr;
  84 //! Convenience typedef
  85 typedef tinyxml2::XMLElement *XMLElementPtr;
  86 //! Convenience typedef
  87 typedef tinyxml2::XMLText *XMLTextPtr;
  88
  89 //! \name Helper functions for XML reading
  90 //! \{
  91
  92 void readEntry(XMLNodePtr node, ReferenceDataEntry *entry);
  93
  94 XMLNodePtr getCDataChildNode(XMLNodePtr node)
  95 {
  96     XMLNodePtr cdata = node->FirstChild();
  97     while (cdata != nullptr &&
  98            cdata->ToText() != nullptr &&
  99            !cdata->ToText()->CData())
 100     {
 101         cdata = cdata->NextSibling();
 102     }
 103     return cdata;
 104 }
 105
 106 bool hasCDataContent(XMLNodePtr node)
 107 {
 108     return getCDataChildNode(node) != nullptr;
 109 }
 110
 111 //! Return a node convertible to text, either \c childNode or its first such sibling.
 112 XMLNodePtr getNextTextChildNode(XMLNodePtr childNode)
 113 {
 114     // Note that when reading, we don't have to care if it is in a
 115     // CDATA section, or not.
 116     while (childNode != nullptr)
 117     {
 118         if (childNode->ToText() != nullptr)
 119         {
 120             break;
 121         }
 122         childNode = childNode->NextSibling();
 123     }
 124     return childNode;
 125 }
 126
 127 //! Return the concatenation of all the text children of \c node, including multiple CDATA children.
 128 std::string getValueFromLeafElement(XMLNodePtr node)
 129 {
 130     std::string value;
 131
 132     XMLNodePtr  childNode = getNextTextChildNode(node->FirstChild());
 133     while (childNode != nullptr)
 134     {
 135         value += std::string(childNode->Value());
 136
 137         childNode = getNextTextChildNode(childNode->NextSibling());
 138     }
 139
 140     if (hasCDataContent(node))
 141     {
 142         // Prepare to strip the convenience newline added in
 143         // createElementContents, when writing CDATA content for
 144         // StringBlock data.
 145         if (value.empty() || value[0] != '\n')
 146         {
 147             GMX_THROW(TestException("Invalid string block in reference data"));
 148         }
 149         value.erase(0, 1);
 150     }
 151
 152     return value;
 153 }
 154
 155 //! Make a new entry from \c element.
 156 ReferenceDataEntry::EntryPointer createEntry(XMLElementPtr element)
 157 {
 158     const char *id = element->Attribute(c_IdAttrName);
 159     ReferenceDataEntry::EntryPointer entry(new ReferenceDataEntry(element->Value(), id));
 160     return entry;
 161 }
 162
 163 //! Read the child entries of \c parentElement and transfer the contents to \c entry
 164 void readChildEntries(XMLNodePtr parentElement, ReferenceDataEntry *entry)
 165 {
 166     XMLElementPtr childElement = parentElement->FirstChildElement();
 167     while (childElement != nullptr)
 168     {
 169         ReferenceDataEntry::EntryPointer child(createEntry(childElement));
 170         readEntry(childElement, child.get());
 171         entry->addChild(move(child));
 172         childElement = childElement->NextSiblingElement();
 173     }
 174 }
 175
 176 //! Return whether \c node has child XML elements (rather than text content).
 177 bool isCompoundElement(XMLNodePtr node)
 178 {
 179     return node->FirstChildElement() != nullptr;
 180 }
 181
 182 //! Read \c element and transfer the contents to \c entry
 183 void readEntry(XMLNodePtr element, ReferenceDataEntry *entry)
 184 {
 185     if (isCompoundElement(element))
 186     {
 187         readChildEntries(element, entry);
 188     }
 189     else if (hasCDataContent(element))
 190     {
 191         entry->setTextBlockValue(getValueFromLeafElement(element));
 192     }
 193     else
 194     {
 195         entry->setValue(getValueFromLeafElement(element));
 196     }
 197 }
 198
 199 //! \}
 200
 201 }       // namespace
 202
 203 //! \cond internal
 204 ReferenceDataEntry::EntryPointer
 205 readReferenceDataFile(const std::string &path)
 206 {
 207     tinyxml2::XMLDocument document;
 208     document.LoadFile(path.c_str());
 209     if (document.Error())
 210     {
 211         const char *errorStr1 = document.GetErrorStr1();
 212         const char *errorStr2 = document.GetErrorStr2();
 213         std::string errorString("Error was ");
 214         if (errorStr1)
 215         {
 216             errorString += errorStr1;
 217         }
 218         if (errorStr2)
 219         {
 220             errorString += errorStr2;
 221         }
 222         if (!errorStr1 && !errorStr2)
 223         {
 224             errorString += "not specified.";
 225         }
 226         GMX_THROW(TestException("Reference data not parsed successfully: " + path + "\n." + errorString + "\n"));
 227     }
 228     XMLElementPtr rootNode = document.RootElement();
 229     if (rootNode == nullptr)
 230     {
 231         GMX_THROW(TestException("Reference data is empty: " + path));
 232     }
 233     if (std::strcmp(rootNode->Value(), c_RootNodeName) != 0)
 234     {
 235         GMX_THROW(TestException("Invalid root node type in " + path));
 236     }
 237
 238     ReferenceDataEntry::EntryPointer rootEntry(ReferenceDataEntry::createRoot());
 239     readEntry(rootNode, rootEntry.get());
 240     return rootEntry;
 241 }
 242 //! \endcond
 243
 244 /********************************************************************
 245  * XML writing
 246  */
 247
 248 namespace
 249 {
 250
 251 //! \name Helper functions for XML writing
 252 //! \{
 253
 254 void createElementAndContents(XMLElementPtr             parentElement,
 255                               const ReferenceDataEntry &entry);
 256
 257 void setIdAttribute(XMLElementPtr element, const std::string &id)
 258 {
 259     if (!id.empty())
 260     {
 261         element->SetAttribute(c_IdAttrName, id.c_str()); // If this fails, it throws std::bad_alloc
 262     }
 263 }
 264
 265 XMLElementPtr createElement(XMLElementPtr parentElement, const ReferenceDataEntry &entry)
 266 {
 267     XMLElementPtr element = parentElement->GetDocument()->NewElement(entry.type().c_str());
 268     parentElement->InsertEndChild(element);
 269     setIdAttribute(element, entry.id()); // If this fails, it throws std::bad_alloc
 270     return element;
 271 }
 272
 273 void createChildElements(XMLElementPtr parentElement, const ReferenceDataEntry &entry)
 274 {
 275     const ReferenceDataEntry::ChildList &children(entry.children());
 276     ReferenceDataEntry::ChildIterator    child;
 277     for (child = children.begin(); child != children.end(); ++child)
 278     {
 279         createElementAndContents(parentElement, **child);
 280     }
 281 }
 282
 283 /*! \brief Handle \c input intended to be written as CDATA
 284  *
 285  * This method searches for any ']]>' sequences embedded in \c input,
 286  * because this must always end a CDATA field. If any are found, it
 287  * breaks the string so that instead multiple CDATA fields will be
 288  * written with that token sequence split across the fields. Note that
 289  * tinyxml2 does not handle such things itself.
 290  *
 291  * This is an edge case that is unimportant for GROMACS refdata, but
 292  * it is preferable to know that the infrastructure won't break.
 293  */
 294 std::vector<std::string> breakUpAnyCdataEndTags(const std::string &input)
 295 {
 296     std::vector<std::string> strings;
 297     std::size_t              startPos = 0;
 298     std::size_t              endPos;
 299
 300     do
 301     {
 302         endPos = input.find("]]>", startPos);
 303         if (endPos != std::string::npos)
 304         {
 305             // We found an embedded CDATA end tag, so arrange to split it into multiple CDATA blocks
 306             endPos++;
 307         }
 308         strings.push_back(input.substr(startPos, endPos));
 309         startPos = endPos;
 310     }
 311     while (endPos != std::string::npos);
 312
 313     return strings;
 314 }
 315
 316 void createElementContents(XMLElementPtr element, const ReferenceDataEntry &entry)
 317 {
 318     // TODO: Figure out if \r and \r\n can be handled without them
 319     // changing to \n in the roundtrip.
 320     if (entry.isCompound())
 321     {
 322         createChildElements(element, entry);
 323     }
 324     else if (entry.isTextBlock())
 325     {
 326         // An extra newline is written in the beginning to make lines align
 327         // in the output xml (otherwise, the first line would be off by the length
 328         // of the starting CDATA tag).
 329         const std::string        adjustedValue = "\n" + entry.value();
 330         std::vector<std::string> cdataStrings  = breakUpAnyCdataEndTags(adjustedValue);
 331         for (auto const &s : cdataStrings)
 332         {
 333             XMLTextPtr textNode = element->GetDocument()->NewText(s.c_str());
 334             textNode->SetCData(true);
 335             element->InsertEndChild(textNode);
 336         }
 337     }
 338     else
 339     {
 340         XMLTextPtr textNode = element->GetDocument()->NewText(entry.value().c_str());
 341         element->InsertEndChild(textNode);
 342     }
 343 }
 344
 345 void createElementAndContents(XMLElementPtr parentElement, const ReferenceDataEntry &entry)
 346 {
 347     XMLElementPtr element = createElement(parentElement, entry);
 348     createElementContents(element, entry);
 349 }
 350
 351 XMLElementPtr createRootElement(XMLDocumentPtr document)
 352 {
 353     XMLElementPtr rootElement = document->NewElement(c_RootNodeName);
 354     document->InsertEndChild(rootElement);
 355     return rootElement;
 356 }
 357
 358 //! \}
 359
 360 }       // namespace
 361
 362 //! \cond internal
 363 void writeReferenceDataFile(const std::string        &path,
 364                             const ReferenceDataEntry &rootEntry)
 365 {
 366     // TODO: Error checking
 367     tinyxml2::XMLDocument     document;
 368
 369     tinyxml2::XMLDeclaration *declaration = document.NewDeclaration(c_VersionDeclarationString);
 370     document.InsertEndChild(declaration);
 371
 372     declaration = document.NewDeclaration(c_StyleSheetDeclarationString);
 373     document.InsertEndChild(declaration);
 374
 375     XMLElementPtr rootElement = createRootElement(&document);
 376     createChildElements(rootElement, rootEntry);
 377
 378     if (document.SaveFile(path.c_str()) != tinyxml2::XML_NO_ERROR)
 379     {
 380         GMX_THROW(TestException("Reference data saving failed in " + path));
 381     }
 382 }
 383 //! \endcond
 384
 385 } // namespace test
 386 } // namespace gmx