This bumps odpic to 3.1.0 - with the current version in unstable,
python*Packages.cx_oracle fails to compile due to some types missing in
our version of odpic:
```
gcc -Wno-unused-result -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -fPIC -DCXO_BUILD_VERSION=7.0.0 -I/nix/store/43lwkzvxwcymshchqhhafr2rnw2kk8ll-odpic-2.4.2/include -I/nix/store/ydk0mfpvn9smcmn72wc9i20slv1d2b79-python3-3.7.2/include/python3.7m -c src/cxoBuffer.c -o build/temp.linux-x86_64-3.7/src/cxoBuffer.o
In file included from src/cxoBuffer.c:17:0:
src/cxoModule.h:372:5: error: unknown type name 'dpiSodaColl'
dpiSodaColl *handle;
^~~~~~~~~~~
src/cxoModule.h:379:5: error: unknown type name 'dpiSodaDb'
dpiSodaDb *handle;
^~~~~~~~~
src/cxoModule.h:386:5: error: unknown type name 'dpiSodaDoc'
dpiSodaDoc *handle;
^~~~~~~~~~
src/cxoModule.h:392:5: error: unknown type name 'dpiSodaDocCursor'
dpiSodaDocCursor *handle;
^~~~~~~~~~~~~~~~
src/cxoModule.h:398:5: error: unknown type name 'dpiSodaOperOptions'
dpiSodaOperOptions options;
^~~~~~~~~~~~~~~~~~
src/cxoModule.h:492:9: error: unknown type name 'dpiSodaColl'; did you mean 'dpiPool'?
dpiSodaColl *handle);
^~~~~~~~~~~
dpiPool
src/cxoModule.h:496:49: error: unknown type name 'dpiSodaDoc'; did you mean 'cxoSodaDoc'?
cxoSodaDoc *cxoSodaDoc_new(cxoSodaDatabase *db, dpiSodaDoc *handle);
^~~~~~~~~~
cxoSodaDoc
src/cxoModule.h:499:9: error: unknown type name 'dpiSodaDocCursor'; did you mean 'cxoSodaDocCursor'?
dpiSodaDocCursor *handle);
^~~~~~~~~~~~~~~~
cxoSodaDocCursor
```
Since #53055 was merged the Makefile for the manual could not be run
correctly as the generated function documentation was included, but
not actually generated.
This adds the necessary generation step by first building the XML file
containing function locations and preserving its store path in a
variable, which is then used both for linking of the locations file
and as a build input for the function docs generator.
This fixes#55014
Comments on conflicts:
- llvm: d6f401e1 vs. 469ecc70 - docs for 6 and 7 say the default is
to build all targets, so we should be fine
- some pypi hashes: they were equivalent, just base16 vs. base32
Pascal Wittmann