This commit does a bulk fix of incorrect annotations (comments) at the
end of namespaces.
The commit was generated by initially running clang-format, and then
from the diff of the result extracting the hunks corresponding to
fixes of namespace comments. The changes being applied and all the
results have been manually reviewed. The source code successfully
builds on macOS.
Further details on the steps below, in case they're of interest to
someone else in the future.
1. Checkout a fresh and up to date version of src/
git pull && git checkout -- src && git status src
2. Ensure there's a suitable .clang-format in place, i.e. with options
to fix the comment at the end of namespaces, including:
FixNamespaceComments: true
SpacesBeforeTrailingComments: 1
and that clang-format is >= 5.0.0, by doing e.g.:
clang-format -dump-config | grep Comments:
clang-format --version
3. Apply clang-format to the source:
clang-format -i $(find src -name "*.cpp" -or -name "*.h")
4. Create and filter out hunks related to fixing the namespace
git diff -U0 src > tmp.patch
grepdiff '^} // namespace' --output-matching=hunk tmp.patch > fix_namespace.patch
5. Filter out hunks corresponding to simple fixes into to a separate patch:
pcregrep -M -e '^diff[^\n]+\nindex[^\n]+\n--- [^\n]+\n\+\+\+ [^\n]+\n' \
-e '^@@ -[0-9]+ \+[0-9]+ @@[^\n]*\n-\}[^\n]*\n\+\}[^\n]*\n' \
fix_namespace.patch > fix_namespace_simple.patch
6. Manually review the simple patch and then apply it, after first
restoring the source.
git checkout -- src
patch -p1 < fix_namespace_simple.path
7. Manually review the (simple) changes and then stage the changes
git diff src
git add src
8. Again apply clang-format and filter out hunks related to any
remaining fixes to the namespace, this time filter with more
context. There will be fewer hunks as all the simple cases have
already been handled:
clang-format -i $(find src -name "*.cpp" -or -name "*.h")
git diff src > tmp.patch
grepdiff '^} // namespace' --output-matching=hunk tmp.patch > fix_namespace2.patch
9. Manually review/edit the resulting patch file to remove hunks for files
which need to be dealt with manually, noting the file names and
line numbers. Then restore files to as before applying clang-format
and apply the patch:
git checkout src
patch -p1 < fix_namespace2.patch
10. Manually fix the files noted in the previous step. Stage files,
review changes and commit.
False positive rate of hints is quite high. Although the includes can be
technically removed (due to other includes) they logically belong to the
header.
Do free free to change the categories. I just did these ones for testing, really.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@31033 a592a061-630c-0410-9148-cb99ea01b6c8
The changes that accomplish this part are in ModuleList.{h,cpp}, configure.py, and the *.module files themselves. This is a format change, and the lyx2lyx is in those files.
By itself, that change would not be major, except for the fact that we do not want the module to be represented in the UI by its filename---e.g., theorems-std---but rather by a descriptive name, such as "Theorems". But that change turns out to be wholly non-trivial. The mechanism for choosing modules was the same as---indeed, was borrowed from---that in GuiCitation: You get a list of modules, and choosing them involves moving strings from one QListView to another. The models underlying these views are just QStringListModels, which means that, when you want to know what modules have been selected, you see what strings are in the "selected" QListView. But these are just the descriptive names, and we can't look up a module by its descriptive name if it's been translated. That, indeed, was the whole point of the change to the new representation.
So, we need a more complicated model underlying the QListView, one that will pair an identifying string---the filename minus the extension, in this case---with each item. This turns out not to be terribly difficult, though it took rather a while for me to understand why it's not difficult. There are two parts:
(i) GuiSelectionManger gets re-written to use any QAbstractListModel, not just a QStringListModel. This actually seems to improve the code, independently.
(ii) We then subclass QAbstractListModel to get the associated ID string, using the Qt::UserRole slot associated with each item to store its ID. This would be almost completely trivial if QAbstractListItem::itemData() included the QVariant associated with this role, but it doesn't, so there are some additional hoops through which to jump.
The new model, a GuiIdListModel, is defined in the files by that name. The changes in GuiSelectionManger.{h,cpp} make it more abstract; the changes in GuiDocument.{h,cpp} adapt it to the new framework.
I've also updated the module documenation to accord with this change.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@22501 a592a061-630c-0410-9148-cb99ea01b6c8
None of the extant modules require any other modules, but the Theorem modules all exclude one another. (See the screenshot.) When I modularize the AMS classes---that is the next task---we'll have requires.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@22456 a592a061-630c-0410-9148-cb99ea01b6c8
Design goal: Allow the use of layout "modules", which are to LaTeX packages as layout files are to LaTeX document classes. Thus, one could have a module that defined certain character styles, environments, commands, or what have you, and include it in various documents, each of which uses a different document class, without having to modify the layout files themselves. For example, a theorems.module could be used with article.layout to provide support for theorem-type environments, without having to modify article.layout itself, and the same module could be used with book.layout, etc.
This patch adds the backend. The ModuleList class holds a list of the available modules, which are retrieved from lyxmodules.lst, itself generated by configure.py. There are two LFUNs available: modules-clear and module-add, which do the obvious thing; you can test by typing these into the minibuffer, along with the name of one of the available modules: URL (a CharStyle), Endnote (a Custom Inset), and---with the spaces---End To Foot (View>LaTeX and look at the user preamble), which are themselves in lib/layouts. There are some others, too, that allow theorems to be added to classes like article and book.
The GUI will come next.
Issues: (i) The configure.py script could be improved. It'd be nice, for example, if it tested for the presence of the LaTeX packages a particular module needs. But this would mean re-working the LaTeX script, and I don't know how to do that. Note that at present, the packages are ignored. This will change shortly. (ii) I've used std::string in LyXModule, following what seemed to be a precedent in TextClass. If some of these should be docstrings, please let me know, and I'll change them. (iii) There is at present no distinction between LaTeX and DocBook modules. Should there be? That is: Should there be modules that are available when the document class is a LaTeX class and others that are available only when it is DocBook? Or should there just be one set of modules? Each module can of course indicate for what it is suitable in its description.
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@19893 a592a061-630c-0410-9148-cb99ea01b6c8
