| | <<<Back 1 day (to 2016/03/23) | 20160324 |
| kraplax | hi everyone | 10:14.58 |
| | i need to create a PDF with defined spot color | 10:15.13 |
| | and i want to embed an image (say TIFF for example) | 10:15.27 |
| | and i want that black-n-white image to represent that spot color channel | 10:15.57 |
| | how do i do that? | 10:16.04 |
| kens | Is ths some kind of general PostScript/PDF question or are you referring to a specific product ? | 10:16.31 |
| sebras | tor8: bmp optimization to review over at sebras/master | 11:20.54 |
| | tor8: still survives testing.sh | 11:21.14 |
| Robin_Watts | sebras: Do you have a set of bmps you're using to test? | 12:03.22 |
| | If you can upload them to casper, I can add them to the test repos and the cluster. | 12:03.36 |
| sebras | Robin_Watts: I do. is it worthwhile to have both valid and corrupt bmps? | 12:47.24 |
| | Robin_Watts: i.e. does the cluster catch segfaults properly? | 12:47.48 |
| | ...should we ever accidentally trigger such. | 12:48.06 |
| Robin_Watts | It does. | 12:48.21 |
| | basically the cluster catches changes in output. | 12:48.43 |
| | So it won't verify files as being correct initially, but it'll spot if anything ever breaks/gets fixed. | 12:49.09 |
| sebras | right. | 12:49.15 |
| | Robin_Watts: ping? | 16:43.26 |
| | Robin_Watts: what does the testsuite need to look like? currenlty I have a file tree of .bmp-files | 16:44.02 |
| | Robin_Watts: but since the .bmp-files come from two git repositories, can I store those there as well? | 16:44.32 |
| | Robin_Watts: does the cluster use a filelist to determine what files to render (and what to ignore)? | 16:44.57 |
| Robin_Watts | I have to add the files to our tests or tests_private git repos. | 16:44.59 |
| | the cluster runs files within directories. | 16:45.19 |
| sebras | Robin_Watts: all files? | 16:45.24 |
| Robin_Watts | So I'd have a 'bmp' directory with your files in. | 16:45.31 |
| sebras | right. | 16:45.36 |
| Robin_Watts | IF you want to split them then bmp/working bmp/broken or something would be OK, but too many dirs is a pain, as it doesn't do recursive searching | 16:46.16 |
| sebras | Robin_Watts: there's a bmps.tar on casper now. | 16:47.06 |
| | Robin_Watts: it has a few directories in them. one per testsuite. | 16:47.26 |
| Robin_Watts | sebras: Ta. will look in a bit. | 16:47.29 |
| sebras | Robin_Watts: and each of those has a good/broken/questionable directory. | 16:47.41 |
| | Robin_Watts: so using these testfiles we achieve a 98.6% coverage of the bmp-code. | 16:48.23 |
| | Robin_Watts: there might be a few paths through the code that aren't covered, but it's not too bad at least. | 16:49.24 |
| | tor8: a few bmp fixes on sebras/master | 17:42.59 |
| malc_ | tor8: pdf_lookup_substitute_font returns const char *, but load_builtin_font takes char * ... | 18:08.29 |
| Robin_Watts | malc_: Does changing pdF_load_builtin_font to take a const solve the problem? | 18:12.21 |
| malc_ | Robin_Watts: load_builtin_font not pdf_load_builtin_font, and no idea, i've just cast the value to remove constness on my side | 18:13.04 |
| | both solutions are icky though | 18:13.22 |
| Robin_Watts | I don't have a load_builtin_font... | 18:13.38 |
| malc_ | Robin_Watts: oh, face palm, it is my own function sorry for the noise | 18:16.07 |
| Robin_Watts | hehe :) | 18:16.17 |
| malc_ | haven't touched this portion of the code in quite a while | 18:16.23 |
| | Forward 1 day (to 2016/03/25)>>> | |
IRC Logs