image/test/reftest/jpeg/reftest.list@ac0c01689b40 (annotated)
image/test/reftest/jpeg/reftest.list
Thu, 15 Jan 2015 15:59:08 +0100
- author
- Michael Schloh von Bennewitz <michael@schloh.com>
- date
- Thu, 15 Jan 2015 15:59:08 +0100
- branch
- TOR_BUG_9701
- changeset 10
- ac0c01689b40
- permissions
- -rw-r--r--
Implement a real Private Browsing Mode condition by changing the API/ABI;
This solves Tor bug #9701, complying with disk avoidance documented in
https://www.torproject.org/projects/torbrowser/design/#disk-avoidance.
| michael@0 | 1 | # JPEG tests |
| michael@0 | 2 | |
| michael@0 | 3 | # Images of various sizes. |
| michael@0 | 4 | == jpg-size-1x1.jpg jpg-size-1x1.png |
| michael@0 | 5 | == jpg-size-2x2.jpg jpg-size-2x2.png |
| michael@0 | 6 | == jpg-size-3x3.jpg jpg-size-3x3.png |
| michael@0 | 7 | == jpg-size-4x4.jpg jpg-size-4x4.png |
| michael@0 | 8 | == jpg-size-5x5.jpg jpg-size-5x5.png |
| michael@0 | 9 | == jpg-size-6x6.jpg jpg-size-6x6.png |
| michael@0 | 10 | == jpg-size-7x7.jpg jpg-size-7x7.png |
| michael@0 | 11 | == jpg-size-8x8.jpg jpg-size-8x8.png |
| michael@0 | 12 | == jpg-size-9x9.jpg jpg-size-9x9.png |
| michael@0 | 13 | == jpg-size-15x15.jpg jpg-size-15x15.png |
| michael@0 | 14 | == jpg-size-16x16.jpg jpg-size-16x16.png |
| michael@0 | 15 | == jpg-size-17x17.jpg jpg-size-17x17.png |
| michael@0 | 16 | == jpg-size-31x31.jpg jpg-size-31x31.png |
| michael@0 | 17 | == jpg-size-32x32.jpg jpg-size-32x32.png |
| michael@0 | 18 | == jpg-size-33x33.jpg jpg-size-33x33.png |
| michael@0 | 19 | # Progressive encoding |
| michael@0 | 20 | == jpg-progressive.jpg jpg-progressive.png |
| michael@0 | 21 | # Grayscale colorspace |
| michael@0 | 22 | == jpg-gray.jpg jpg-gray.png |
| michael@0 | 23 | # CMYK colorspace |
| michael@0 | 24 | == jpg-cmyk-1.jpg jpg-cmyk-1.png |
| michael@0 | 25 | == jpg-cmyk-2.jpg jpg-cmyk-2.png |
| michael@0 | 26 | # This intermittently fails on Android due to async image decoding (bug #685516) |
| michael@0 | 27 | # Sometimes the image decodes in time and the test passes, other times the image |
| michael@0 | 28 | # appears blank and the test fails. This only seems to be triggered since the |
| michael@0 | 29 | # switch to 24-bit colour (bug #803299). |
| michael@0 | 30 | random-if(Android) == jpg-srgb-icc.jpg jpg-srgb-icc.png |
| michael@0 | 31 | |
| michael@0 | 32 | # webcam-simulacrum.mjpg is a hand-edited file containing red.jpg and blue.jpg, |
| michael@0 | 33 | # concatenated together with the relevant headers for |
| michael@0 | 34 | # multipart/x-mixed-replace. Specifically, with the headers in |
| michael@0 | 35 | # webcam-simulacrum.mjpg^headers^, the web browser will get the following: |
| michael@0 | 36 | # |
| michael@0 | 37 | # HTTP 200 OK |
| michael@0 | 38 | # Content-Type: multipart/x-mixed-replace;boundary=BOUNDARYOMG |
| michael@0 | 39 | # |
| michael@0 | 40 | # --BOUNDARYOMG\r\n |
| michael@0 | 41 | # Content-Type: image/jpeg\r\n |
| michael@0 | 42 | # \r\n |
| michael@0 | 43 | # <contents of red.jpg> (no newline) |
| michael@0 | 44 | # --BOUNDARYOMG\r\n |
| michael@0 | 45 | # Content-Type: image/jpeg\r\n |
| michael@0 | 46 | # \r\n |
| michael@0 | 47 | # <contents of blue.jpg> (no newline) |
| michael@0 | 48 | # --BOUNDARYOMG--\r\n |
| michael@0 | 49 | # |
| michael@0 | 50 | # (The boundary is arbitrary, and just has to be defined as something that |
| michael@0 | 51 | # won't be in the text of the contents themselves. --$(boundary)\r\n means |
| michael@0 | 52 | # "Here is the beginning of a boundary," and --$(boundary)-- means "All done |
| michael@0 | 53 | # sending you parts.") |
| michael@0 | 54 | HTTP == webcam-simulacrum.mjpg blue.jpg |
| michael@0 | 55 | pref(image.mozsamplesize.enabled,true) fuzzy(21,256) == jpg-size-32x32.jpg#-moz-samplesize=2 jpg-size-16x16.png |
| michael@0 | 56 | pref(image.mozsamplesize.enabled,true) fuzzy(92,16) == jpg-size-32x32.jpg#-moz-samplesize=8 jpg-size-4x4.png |
