summaryrefslogtreecommitdiff
path: root/solenv/gdb
AgeCommit message (Collapse)Author
2018-04-04Fix SfxItemSet pretty printer after some uniqueptr conversionsKatarina Behrens
just whyyyyyyy are you people doing this, why do you gratuitously break things others rely upon for productivity?!?? I don't give a F about uniqueptrs, all I want is to be able to print this bloody SfxItemSet. It's frustrating AF Change-Id: I0b6110c1806c31a3c098fb9dc73883047191c7f9 Reviewed-on: https://gerrit.libreoffice.org/52362 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Katarina Behrens <Katarina.Behrens@cib.de>
2018-03-14fix pretty printing of Fraction betterNoel Grandin
work-around gdb bug, and deal with different representations of std::unique_ptr in different versions of libc++. Change-Id: I02a1f49c07dbcd70e13ef6be48e20b510bf5e3fa
2018-03-14Adapt FractionPrinterStephan Bergmann
...to 048a32d903c8cc81a8836c7c7209b92e0bfd6cd7 "use unique_ptr for pImpl in tools/", exploiting intimate knowledge of libstdc++ internals as was already done in 51bb68928933bf5d72efb9193f1be6af6e72c80f "fix BigPtrArray pretty printer". Due to <https://sourceware.org/bugzilla/show_bug.cgi?id=22968> "ptype does not find inner C++ class type without -readnow", at least for me only works when gdb is invoked with -readnow. Change-Id: I365733f3c661eea2f19a9b85015aa5d50e63e6a4
2017-11-08gdb pretty-printers: avoid segfauls in B2DPolygonPrinterMichael Stahl
gdb 8.0.1 tends to sefault while evaluating the getB2DPoint() calls; it passes some obviously wrong index instead of 0. Also, add a children method to B2DPolyPolygonPrinter. Change-Id: Ifbf52ad384d1f60b26ee95f87405eff2c6f2388a
2017-09-29gdb pretty-printers: fix StringPrinterHelper for gdb 8.0Michael Stahl
Latest gdb release "lazy_string" validates the array size: Traceback (most recent call last): File "/work/lo/master/solenv/gdb/libreoffice/util/string.py", line 29, in to_string return self.make_string(data, self.encoding, len) File "/work/lo/master/solenv/gdb/libreoffice/util/string.py", line 66, in make_string return data.lazy_string(encoding, length) gdb.error: Length is larger than array size. rtl_uString has "sal_Unicode buffer[1];", which is a lie as the real size is the same as "length". Taking the address of "buffer" appears to avoid the exception. Change-Id: I85710b1adfae584ba09c8d517e9b49b290e79d8a
2017-09-15solenv: fix GDB pretty-printers for SwNodes yet againMichael Stahl
Change-Id: I1eb1efedb98979b692874ae0ec48120cb5a38188
2017-07-29Fix no longer valid commentPranav Kant
This was changed in d72aad218c9737fb19d1a835b03c13b7107a96c0 Change-Id: Ie93f603c82669ca612fbf9635a62c50e63702a85
2017-07-28Fix scheduler GDB pretty printerJan-Marek Glogowski
mbDelete is gone, as we simply check for mpTask. Change-Id: I9d451f2445fb6b978d177b45d75abde2595c5fac
2017-07-13fix BigPtrArray pretty printerNoel Grandin
after commit 64079184cc06444e9dfdb55a5c66829e929d4a59 "improve useuniqueptr loplugin to find arrays" Change-Id: I29841af65fcb0c1f92e52d9aeee164328bbb4663
2017-07-13Run Idle tasks immediatlyJan-Marek Glogowski
There is really no reason to wait a millisecond for an idle. Change-Id: I7665d5f2e7d6ba3e01290a692bbc8e42c36b9986
2017-07-13Introduce a scheduler stackJan-Marek Glogowski
While the stack removes all invoked tasks from the queue, which actively removes it from scheduling, it also helps to faster handle nested calls, as we don't have to look for the previous position to move the task to the end of the queue for the round robin. Change-Id: I358cf2492e9630f67685a2b780509edb56691830
2017-07-07re-add some gdb BigPtrArray pretty printingNoel Grandin
After commit c6902761d797253cda8b3f71f102c66108585e24 "Revert "use std::vector in BigPtrArray"" <mst_> noelgrandin: you reverted both gdb changes but only one sw change? <noelgrandin> mst_, I thought your gdb change only appliled to the std::vector? <mst_> noelgrandin: there was another std::something or other change there, one was mvInfo and the other mvData <noelgrandin> mst_, sigh, will fix that Change-Id: I3059d29b1a559eb3077f5d713be011e15e887ee0
2017-07-07Revert "use std::vector in BigPtrArray"Noel Grandin
which is causing crashes in the crashtesting in ooo119635-3.docx and ooo119568-2.docx It is definitely some kind of use-after-free error, but the compress and delete logic for BigPtrArray is too hairy for me to debug right now. This reverts commit 1eee0abd459a508a6dcf9e71cbf2c1be3725faa7. Also revert commit 4f743419a04375160437a910254c45dea396f70d "gdb pretty-printers: fix BigPtrArrayPrinter after recent std::isation" Change-Id: Id870876432a060f9347aafb43bf0df692ea24464 Reviewed-on: https://gerrit.libreoffice.org/39684 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2017-06-28gdb pretty-printers: fix BigPtrArrayPrinter after recent std::isationMichael Stahl
Change-Id: Ie98f080fbb0efb807dcb2fb7893811f68e831a8f
2017-04-04Missing str(...) when an arg is itself a css::uno::Sequence<...>Stephan Bergmann
Change-Id: I54529e7086014a2feba89eb73f3b368d36f758b8
2017-01-23Revert "used std::map in SfxItemSet"Noel Grandin
This reverts commit 2757ee9fe610e253e4ccc37423fa420004d0f388. Besides causing a performance regression, I now notice that there is code in SW that relies on iterating over two different SfxItemSet's in parallel, and assumes that missing items are returned as nullptr, which is not the case for my std::map based change. Change-Id: I2b1110350fe4c4b74e5508558e9661ef1e1a103e
2017-01-17Change Idle to be a Timer subclassJan-Marek Glogowski
Drops a lot of duplicated code, as Idle is just a convenience class for instant, mostly low priority timers. Change-Id: I847592e92e86d15ab1cab168bf0e667322e48048
2017-01-17Refactor Scheduler to add Task classJan-Marek Glogowski
Moves all the "task-specific" stuff into a Task class and just keeps the "real" static Scheduler functions in the original Scheduler class. Change-Id: I9eb02d46e2bcf1abb06af5bab1fa0ee734d1984c
2017-01-17tdf#97087 GDB pretty print the Scheduler task listJan-Marek Glogowski
In addition to the GDB pretty printer, this annotates a lot more Timers and Idles. Change-Id: I5b93fab02161b23bb753e65ef92643a04fb0789c
2017-01-17used std::map in SfxItemSetNoel Grandin
instead of naked array SfxItemIter ended up needing to take copies of stuff because various code likes to iterate over the items and delete items inside the loop. The gdb pretty printer is no longer quite as pretty as it was before, but it still prints useful info. Change-Id: I59b07ea42f6b1c74798a15402970b9dbd8233dbe
2016-12-20gdb pretty printers: fix the SwXTextCursor::Impl one even moreMichael Stahl
Turns out that commit 30f3315d52df22716c086836c41465a6c4cb98b5 has removed the SwClient there a long time ago, and the SwUnoCrsr was also renamed... Change-Id: I92932130c5adbea530c39557932efce70c9f75e7
2016-12-20gdb pretty printers: fix BigPtrArray printer for member renameMichael Stahl
Change-Id: Ia9da0c0931a01b2c99c24420a9ba603b47a711c1
2016-12-20gdb pretty printers: fix SwXTextCursor::Impl printer for SwClient renameMichael Stahl
Change-Id: Ief136f57ce60bde1faa0603961aa775967621fbf
2016-12-20gdb pretty printers: fix SwPaM printer for Ring member renamingMichael Stahl
Change-Id: I605e4e972468165e7f5d21681b6e7eeb228dfdc9
2016-12-14solenv: adapt SwPositionPrinter to BigPtrEntry renamingMichael Stahl
Change-Id: I096211333c563b501627fef0dbdec8c3c98d501a
2016-10-23Fix typosAndrea Gelmini
Change-Id: Ib7b17f85c7b6a1937c3f6e1617ceec58074643b4 Reviewed-on: https://gerrit.libreoffice.org/30040 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: jan iversen <jani@documentfoundation.org> Tested-by: jan iversen <jani@documentfoundation.org>
2016-07-08Resolves: tdf#100452 class Date full (BCE,CE) proleptic Gregorian calendarEike Rathke
... implementing signed years with year 0 gap. Date(31,12,-1) last day BCE Date(1,1,1) first day CE New class Date member functions: * AddYears(sal_Int16) to be used instead of aDate.SetYear(aDate.GetYear()+sal_Int16) to handle year 0 gap. * convenience GetNextYear() to be used insted of GetYear()+1 * convenience GetPrevYear() to be used insted of GetYear()-1 * AddMonths(sal_Int32) * operator=(const css::util::Date&) New class DateTime member functions: * operator=(const css::util::DateTime&) Made some conversion ctors explicit, specifically Date(sal_Int32) Adapted hopefully all places that used a sal_uInt16 year to use sal_Int16 where appropriate. Eliminated some quirks in date handling found on the fly. Added era handling to i18npool icu calendar setting interface, which missing was responsible for 0001-01-01 entered in Calc being set as -0001-01-01, hence subtracting one day resulted in -0002-12-31. Change-Id: I77b39fba9599ebd5067d7864f6c9ebe01f6f578f Reviewed-on: https://gerrit.libreoffice.org/27049 Reviewed-by: Eike Rathke <erack@redhat.com> Tested-by: Jenkins <ci@libreoffice.org>
2016-06-16Improve output of UnoAnyPrinterStephan Bergmann
Change-Id: I1461264fe7f951c9ecec777c6217d0acfb7e4254
2016-01-02GDB: pretty-print more sw::mark::MarkBase classesJan-Marek Glogowski
Generalize the UnoMark pretty printer to actually print many objects inheriting from sw::mark::MarkBase. Also adds the mark name to the output. Change-Id: Idaec06ba448702ee3a47b474736df954646c0300
2015-12-07loplugin:mergeclasses OOXMLPropertyImpl into OOXMLPropertyNoel Grandin
Change-Id: I1447cfc76c529332450c90a21b6525d3103fe852
2015-12-07loplugin:mergeclasses, merge OOXMLPropertyImpl into OOXMLPropertyNoel Grandin
Change-Id: I73eb7e664c0a53f135c06c3d8ea450bd9493c780
2015-07-30gdb pretty printers for boost can't iterate with Python 3Michael Stahl
Change-Id: Ie2d1cb7312de6f14a5c6de81eefd7a00be6f75c0
2015-07-27gdb pretty printer for SwNodes fails on Python 3 due to "unicode"Michael Stahl
Change-Id: I8e4983a91d4f97a2a20fbeed89d4e0f186c35fad
2015-06-15sw: prefix members of SwNodeIndexMichael Stahl
Change-Id: Ia31c7a91bf78b967e70d288374d9e21439c99221
2015-05-26gdb: SwTxtNode -> SwTextNodeMiklos Vajna
Fixes the unexpected ' ~DeletedNode' lines in the output. Change-Id: I1f59c2cd986addd08e632d0bc1cc53b33048db77
2015-04-08solenv: there are no more SvArrays in need of pretty printingMichael Stahl
Change-Id: Ic4d59228f3295e8400bcacf637ff4def7669b775
2015-04-08solenv: add GDB pretty printer for SfxItemSetMichael Stahl
Change-Id: Ibd74c58efa7f1a24de409820655fb98b4fc13df3
2015-01-29gdb pretty printers: fix the sw::mark::IMark and SwModify printersMichael Stahl
If these are called on a value that is of a different type than what is checked in the children() method, nothing is printed except exceptions about non-existent _iterator. GDB can figure out the dynamic type itself and call the pretty printer only if it's actually a UnoMark so don't register it for IMark. Same thing for SwUnoCrsr, there is already a class for it. Change-Id: I00238f1e3f01741d607eec9d47671302b858b965
2015-01-29gdb pretty printers: remove obsolete tools container printersMichael Stahl
Change-Id: Ieb1e104e8147376405e90ecada308fee05321dcd
2015-01-27Adapt FractionPrinterStephan Bergmann
Change-Id: Ibc6302ea52458874f748f3ec2757bb60ad742fce
2014-11-10Do not deref (potentially dangling) weak_ptrStephan Bergmann
Change-Id: Ic9694aa8bd85fe0548b8f5fc5f52b4bbd6dce1cf
2014-11-10Adapt FractionPrinterStephan Bergmann
Change-Id: Ic297f14ea1bf5f3fa69c80a34439b5d13fc84346
2014-10-24More useful pretty printing of invalid fractionsStephan Bergmann
Change-Id: I5e118182e146645ee8546c69edf21834117a1129
2014-08-20gdb: print the SwNodeIndex properly if there are multiple blocksMichael Stahl
Change-Id: I178d365e71ef3377b1a2cfc115bb297640161efe
2014-01-31Remove UNOIDL "array" and "union" vaporware remnantsStephan Bergmann
...and deprecate what cannot be removed for compatibility. Change-Id: I1ea335af775b867b468b8285113631167729a92a
2013-11-15Add Python 3 compatibility to GDB pretty printers.Jan-Marek Glogowski
GDB on *buntu is linked against Python 3.3, which has many incompatibilities to Python 2, resulting in broken code. This patch uses the Python six library as a compatibility layer. Change-Id: Icb4cc54a1d05afb119376bb5e1430c91cb794d08 Reviewed-on: https://gerrit.libreoffice.org/6688 Reviewed-by: Michael Stahl <mstahl@redhat.com> Tested-by: Michael Stahl <mstahl@redhat.com>
2013-11-15Import Python six v1.4.1.Jan-Marek Glogowski
Python six just consists of a single Python file, so this includes the file as a convenience. Change-Id: I98ed50fba657013f5f3f12a4d13b3ba5558bfa2e Reviewed-on: https://gerrit.libreoffice.org/6687 Reviewed-by: Michael Stahl <mstahl@redhat.com> Tested-by: Michael Stahl <mstahl@redhat.com>
2013-10-22Resolves: fdo#38838 remove UniStringCaolán McNamara
hammer silver nails into coffin and bury in concrete Change-Id: I3fda2ff47738bb33793adab97faba2d439ac9a28
2013-07-28Adapt gdb python pretty-printer to time nano-second precision API changeLionel Elie Mamane
Change-Id: I416d98a298f00ae445a480c738a47758544d317c
2013-06-13Only print the rtl::Reference's pointerStephan Bergmann
...but do not pretty-print its dereferenced value. This is in line with the handling of css::uno::Reference, and avoids gdb Python exceptions when trying to print uninitialized rtl::Reference instances (which can e.g. happen during a "backtrace full"). Change-Id: I9a3c0a6441cf23cba748183226832d2ba23bd531
generated by cgit (git 2.34.1) at 2025-01-29 04:56:57 +0000