tiny cluster fix for latest gstreamer

This commit is contained in:
Stefan Salewski 2021-06-13 10:08:02 +02:00
commit 5cf3c9f0d3
3 changed files with 7 additions and 5 deletions

View file

@ -1,6 +1,6 @@
= High level GTK4 and GTK3 bindings for the Nim programming language = High level GTK4 and GTK3 bindings for the Nim programming language
//(c) Stefan Salewski //(c) Stefan Salewski
//Version 0.9.0 //Version 0.9.2
:experimental: :experimental:
:imagesdir: http://ssalewski.de/tmp :imagesdir: http://ssalewski.de/tmp
//:source-highlighter: pygments //:source-highlighter: pygments
@ -29,8 +29,10 @@ with code inserted from files directly.
NOTE: This work is partly based on earlier works of J. Mansour and has been supported by A. Rumpf, E. Bassi and other _Nim_ and _GTK/Gnome_ developers. NOTE: This work is partly based on earlier works of J. Mansour and has been supported by A. Rumpf, E. Bassi and other _Nim_ and _GTK/Gnome_ developers.
The `combinatorics` module was kindly provided by R. Behrends. The `combinatorics` module was kindly provided by R. Behrends.
NOTE: We are currently preparing version 0.9.2 of the gintro Nim bindings which fixes the latest gstreamer issue -- you may install it
with nimble install gintro@#head.
NOTE: This is version 0.9.1 of the gintro library. It is mostly a plain fix for issue https://github.com/StefanSalewski/gintro/issues/133. NOTE: This version 0.9.1 is mostly a plain fix for issue https://github.com/StefanSalewski/gintro/issues/133.
For GObject proc parameters with direction in and transfer full we have to avoid that the Nim memory management destroys the GObject when the Nim proxy object is destroyed. Current fix was adding a plain ".ignoreFinalizer = true" for that case, which works in most cases. But maybe a better fix would be to ref() the gtk object instead. Maybe that makes no difference for gobjects, but it can make a difference for entities like GtkExpression which are used further, see issue For GObject proc parameters with direction in and transfer full we have to avoid that the Nim memory management destroys the GObject when the Nim proxy object is destroyed. Current fix was adding a plain ".ignoreFinalizer = true" for that case, which works in most cases. But maybe a better fix would be to ref() the gtk object instead. Maybe that makes no difference for gobjects, but it can make a difference for entities like GtkExpression which are used further, see issue
https://github.com/StefanSalewski/gintro/issues/137. But we will leave that for next version 0.9.2. https://github.com/StefanSalewski/gintro/issues/137. But we will leave that for next version 0.9.2.

View file

@ -1,6 +1,6 @@
# Package # Package
version = "0.9.1" version = "0.9.2"
author = "Stefan Salewski" author = "Stefan Salewski"
description = "High level GObject-Introspection based GTK4/GTK3 bindings" description = "High level GObject-Introspection based GTK4/GTK3 bindings"
license = "MIT" license = "MIT"

View file

@ -1,6 +1,6 @@
# High level gobject-introspection based GTK4/GTK3 bindings for the Nim programming language # High level gobject-introspection based GTK4/GTK3 bindings for the Nim programming language
# nimpretty --maxLineLen:130 gen.nim # nimpretty --maxLineLen:130 gen.nim
# v 0.9.0 2021-JUN-01 # v 0.9.2 2021-JUN-13
# (c) S. Salewski 2018, 2019, 2020, 2021 # (c) S. Salewski 2018, 2019, 2020, 2021
# usefull for finding death code: # usefull for finding death code:
@ -3750,7 +3750,7 @@ proc cstringArrayToSeq*(s: ptr cstring): seq[string] =
supmod4.cut(supmodpos4) supmod4.cut(supmodpos4)
delayedSyms = unp delayedSyms = unp
block myb: block myb:
for a in 1 .. 2: # process clustered symbols into one single type section for a in 1 .. 3: # process clustered symbols into one single type section
if delayedSyms.len > a: if delayedSyms.len > a:
suppressType = true suppressType = true
methodBuffer = newStringStream() methodBuffer = newStringStream()