For some reason gitlab doesn't like it.

It worked with the line

> for ext in exe msi 7z zip; do mv ${SHORTARCH}/vlc-*-dev-*.${ext} nightlies/$(basename ${SHORTARCH}/vlc-*-dev-*.${ext} | sed "s/\.${ext}/-${CI_COMMIT_SHORT_SHA}\.${ext}/"); done

It should work with the line

> for ext in exe msi 7z zip; do mv ${SHORTARCH}/vlc-*-dev-*.${ext} nightlies/$(basename ${SHORTARCH}/vlc-*-dev-*.${ext} | sed "s/\.${ext}/-${CI_COMMIT_SHORT_SHA}\.${ext}/") || echo "vlc-*.${ext} not found!"; done

No functional changes. Just bring some consistency between the call to make it
easier to refactor the code if needed.

No functional changes. Just makes it easier to refactor the code if needed.

Compiler may generate better code and we really want to skip exactly that
amount. This will also bring some consistency with the other
vlc_stream_Read(NULL) callers.

The end of file is probably reached. This is the same kind of error checked in
the vlc_stream_Peek() call below.

It's either 0 or 3.

It's filled with uint16_t values or constants.

It can never be negative as it's only growing from 0. And we don't want to skip
backward.

It should have the same range as i_pos.

It's filled with uint32_t values that cannot be higher than UINT32_MAX or
constants.

It's filled with a uint32_t value that is checked not to underflow from the
substraction.

It's filled with uint32_t values or constants.

Regression from 63b45cfc

François Cartegnie authored Oct 17, 2022 and Steve Lhomme committed Nov 01, 2022

Parsing bitmapinfoheader resets to MS fourcc.
Demuxer's codec conditionals cannot match.

François Cartegnie authored Oct 16, 2022 and Steve Lhomme committed Nov 01, 2022

was always using the previous chunk size

refs #27425

François Cartegnie authored Oct 13, 2022 and Steve Lhomme committed Nov 01, 2022

Avoids sending [a single] montrosample[s]

refs #27393

Steve Lhomme authored Oct 26, 2022 and Steve Lhomme committed Nov 01, 2022

It make similarities more obvious.