Fix xmp exporter #3964
Fix xmp exporter #3964
Conversation
Adding a new exporter for xmp export in a .xmp file.
…for generating a single .xmp for every bib entry
Deleted the -xmp cli option, because the pdf exporter has now the same functionality and is embedded within the common cli exporter infrastructure.
johannes-manner
added
the
status: ready-for-review
Due to the elimination of the ?xpacket tags, the number of lines in the correponding files, generated by the xmp exporter, changed.
| @@ -64,8 +67,16 @@ public void export(BibDatabaseContext databaseContext, Path file, Charset encodi | |||
| } | |||
|
|
|||
| private void writeBibToXmp(Path file, List<BibEntry> entries, Charset encoding) throws IOException { | |||
| String xmpContent = XmpUtilWriter.generateXmpString(entries, this.xmpPreferences); | |||
There was a problem hiding this comment.
Can we add a second method there? generateXmpStringWithoutXmpDeclaration? Maybe rename to current functio to generateXmpStringWithXmpDeclaration. The term "XmpDeclaration" stems from "XmlDeclaration". See https://wiki.selfhtml.org/wiki/XML/Regeln/XML-Deklaration.
| @@ -61,12 +63,12 @@ private void extractAuthor() { | |||
| } | |||
|
|
|||
| /** | |||
| * Year + Month in BibTex - Date in DublinCore is a combination of year and month information | |||
| * Year in BibTex - Date in DublinCore is only the year information, because dc interprets empty months as January. | |||
There was a problem hiding this comment.
I don't get this change? The old code from line 81 on seems to extract the month perfectly. In case no month is there, the fallback was and still seems to be January.
There was a problem hiding this comment.
yes the fallback is January, but I thought it's better to not include default values like January....
In my opinion, data consistency is somehow corrupted...
| relationships.stream() | ||
| .filter(isBibTeXElement) | ||
| .forEach(splitBibTeXElement); | ||
|
|
| int i = r.indexOf('/'); | ||
| if (i != -1) { | ||
| bibEntry.setField(r.substring(0, i), r.substring(i + 1)); | ||
| Predicate<String> isBibTeXElement = s -> s.startsWith("bibtex/"); |
There was a problem hiding this comment.
Move this down just after List<String relationships =.
| Consumer<String> splitBibTeXElement = s -> { | ||
| // split solution is complicated, because some fields contains url etc. | ||
| String temp = s.substring("bibtex/".length()); | ||
| int i = temp.indexOf('/'); |
There was a problem hiding this comment.
Please add an example why you need this substring. It is clear from the code that it is bibtex/key/value, but it should be documented somewhere (and referenced from here).
There was a problem hiding this comment.
You are right with the schema bibtex/key/value.
In a former approach I used String#split, but in some cases the value is an url or doi etc.
So the value in this case can contain "/" and therefore the substring is the easiest way to handle this. Documenting the url problem would be beneficial for future changes.
|
|
||
| // only for month field - override value | ||
| // workaround, because the date value of the xmp component of pdf box is corrupted | ||
| if ("month".equals(key)) { |
There was a problem hiding this comment.
We are in the special bibtex mode. So, we hit bibtex/month/#mar# in this case, didn't we? Did we hit bibtex/month/mar? In the former case, I would assume, the export is corrupted. In the latter case, I don't know, what's happening. In all cases, it strange that the date value of PdfBox is corrupt; we are in our custom bibtex mode 😇
koppor
removed
the
status: ready-for-review
Tested the changes manually. Results are now as discussed with @koppor.
|
Thank you for the work! I got a minor comment on the "January, 1st" hack: There should be a comment on that. Or a MADR linked via |
* upstream/master: (47 commits) Fix xmp exporter (#3964) Update JUnit from 5.2.0-M1 -> 5.2.0-RC1 Update xmlunit from 2.5.1 -> 2.6.0 Update mockito-core from 2.18.0 -> 2.18.3 Fixieee (#3970) Fix IEEE Fetcher by enabling cookie support (#3968) Update flowless from 0.6 -> 0.6.1 Update wiremock from 2.16.0 -> 2.17.0 Fix ebook.de (#3967) LOC test #3854 Fix arxiv tests New translations JabRef_en.properties (French) (#3963) Upgrade modernizer plugin New Crowdin translations (#3962) Fix test Improve test New Crowdin translations (#3961) New translations JabRef_en.properties (French) (#3960) Remove brackets before checking equality Replace all IEEE URLs with https #3930 (#3944) ... # Conflicts: # build.gradle # src/main/java/org/jabref/gui/BasePanel.java # src/main/java/org/jabref/gui/actions/CleanupAction.java # src/main/java/org/jabref/gui/collab/ChangeScanner.java # src/main/java/org/jabref/gui/exporter/ExportAction.java # src/main/java/org/jabref/gui/fieldeditors/LinkedFilesEditorViewModel.java # src/main/java/org/jabref/gui/openoffice/DetectOpenOfficeInstallation.java # src/main/java/org/jabref/gui/openoffice/OpenOfficeSidePanel.java # src/main/java/org/jabref/gui/preftabs/PreferencesDialog.java
Solves koppor#338.
Changed the behavior of the xmp exporter. Remove the ?xpacket tags with xmp metadata in dublin core format to enable users to use the .xmp files, exported by JabRef, with the xmpincl LaTeX package.
Also adressed the month parsing problem, implemented a workaround and created an issue for the date section within xmpbox (https://issues.apache.org/jira/browse/PDFBOX-4196).