Jaren Stangret wrote:> Testing the metadata-handling properties of flac-to-flac encoding... > case00a... Files case00a-expect.meta and out.meta differ > ERROR: metadata does not match expected case00a-expect.meta > make: *** [fullcheck] Error 1Last night I did a git bisect on this issue and this error arose when I changed the version from "1.2.1" to "1.3.0pre1". Obvioulsy the test shouldn't have failed just because the version string changed :-). Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/
> Jaren Stangret wrote: > >> Testing the metadata-handling properties of flac-to-flac encoding... >> case00a... Files case00a-expect.meta and out.meta differ >> ERROR: metadata does not match expected case00a-expect.meta >> make: *** [fullcheck] Error 1 > > Last night I did a git bisect on this issue and this error arose > when I changed the version from "1.2.1" to "1.3.0pre1". Obvioulsy > the test shouldn't have failed just because the version string > changed :-).The metadata block contains the vendor string (aka libFLAC version number). I bet case00a-expect.meta still has 1.2.1. -Ben
>> Jaren Stangret wrote: >> >>> Testing the metadata-handling properties of flac-to-flac encoding... >>> case00a... Files case00a-expect.meta and out.meta differ >>> ERROR: metadata does not match expected case00a-expect.meta >>> make: *** [fullcheck] Error 1 >> >> Last night I did a git bisect on this issue and this error arose >> when I changed the version from "1.2.1" to "1.3.0pre1". Obvioulsy >> the test shouldn't have failed just because the version string >> changed :-). > > The metadata block contains the vendor string (aka libFLAC version > number). I bet case00a-expect.meta still has 1.2.1.Slight correction. It isn't checking against the vendor string. However, it is checking against the length of the Vorbis Comments Metadata Block. the "pre1" adds 4 bytes and the test fails. -Ben