Rename test/testdata to test/data

This commit is contained in:
Nigel Tao committed 2018-03-16 10:48:33 +11:00
1 parent bbc3d34e1b
commit d1399fac19
107 files changed
+110 -114

No files matched your search

+6 -6
View File
@@ -21,9 +21,9 @@ Wrangling those formats that build on deflate (gzip, zip and zlib) is not
provided by this package. For zlib, look at the `std/zlib` package instead. The
other formats are TODO.
For example, look at `test/testdata/romeo.txt*`. First, the uncompressed text:
For example, look at `test/data/romeo.txt*`. First, the uncompressed text:
$ xxd test/testdata/romeo.txt
$ xxd test/data/romeo.txt
00000000: 526f 6d65 6f20 616e 6420 4a75 6c69 6574 Romeo and Juliet
00000010: 0a45 7863 6572 7074 2066 726f 6d20 4163 .Excerpt from Ac
etc
@@ -32,7 +32,7 @@ For example, look at `test/testdata/romeo.txt*`. First, the uncompressed text:
The raw deflate encoding:
$ xxd test/testdata/romeo.txt.deflate
$ xxd test/data/romeo.txt.deflate
00000000: 4d53 c16e db30 0cbd f32b d853 2e46 0e3d MS.n.0...+.S.F.=
00000010: 2e87 20ed 0234 c5ba 0049 861e 861d 149b .. ..4...I......
etc
@@ -45,7 +45,7 @@ C string name of the original, uncompressed file, "romeo.txt", amongst other
data. The footer contains a 4 byte CRC32 checksum and a 4 byte length of the
uncompressed file (0x3ae = 942 bytes).
$ xxd test/testdata/romeo.txt.gz
$ xxd test/data/romeo.txt.gz
00000000: 1f8b 0808 26d8 5d59 0003 726f 6d65 6f2e ....&.]Y..romeo.
00000010: 7478 7400 4d53 c16e db30 0cbd f32b d853 txt.MS.n.0...+.S
etc
@@ -56,7 +56,7 @@ The zlib format wraps a 2 byte header and 4 byte footer around the raw deflate
data. The footer contains a 4 byte Adler32 checksum. TODO: move this to
std/zlib/README.md.
$ xxd test/testdata/romeo.txt.zlib
$ xxd test/data/romeo.txt.zlib
00000000: 789c 4d53 c16e db30 0cbd f32b d853 2e46 x.MS.n.0...+.S.F
00000010: 0e3d 2e87 20ed 0234 c5ba 0049 861e 861d .=.. ..4...I....
etc
@@ -66,7 +66,7 @@ std/zlib/README.md.
# Wire Format Worked Example
Consider `test/testdata/romeo.txt.deflate`. The relevant spec is RFC 1951.
Consider `test/data/romeo.txt.deflate`. The relevant spec is RFC 1951.
offset xoffset ASCII hex binary
000000 0x0000 M 0x4D 0b_0100_1101
+10 -10
View File
@@ -7,7 +7,7 @@ specification](https://www.w3.org/Graphics/GIF/spec-gif89a.txt).
# Wire Format Worked Example
Consider `test/testdata/bricks-nodither.gif`.
Consider `test/data/bricks-nodither.gif`.
offset xoffset ASCII hex binary
000000 0x0000 G 0x47 0b_0100_0111
@@ -206,10 +206,10 @@ Significant Bits) and MSB (Most Significant Bits) first. The GIF format uses
LSB first. The PDF and TIFF formats use MSB first, but are not discussed here.
It is not an official format, but the
`test/testdata/bricks-nodither.indexes.giflzw` file contains the extracted
"Pixel Data" payload from `test/testdata/bricks-nodither.gif`, preceded by the
one byte `log2(literal_width)`. Deriving a separate file, as a separate,
contiguous stream, makes it easier to discuss the LZW compression format.
`test/data/bricks-nodither.indexes.giflzw` file contains the extracted "Pixel
Data" payload from `test/data/bricks-nodither.gif`, preceded by the one byte
`log2(literal_width)`. Deriving a separate file, as a separate, contiguous
stream, makes it easier to discuss the LZW compression format.
offset xoffset ASCII hex binary
000000 0x0000 . 0x08 0b_0000_1000
@@ -284,8 +284,8 @@ depending on `max`. For example, if `max` is in the range [256, 511] then the
next code takes 9 bits, if `max` is in the range [512, 1023] then `width ==
10`, and so on. For GIF, when a code spans multiple bytes, codes are formed
Least Significant Bits first. For the
`test/testdata/bricks-nodither.indexes.giflzw` example, `l2lw == 8` and so
`width` starts at 9 bits. Printing the bits on byte boundaries give:
`test/data/bricks-nodither.indexes.giflzw` example, `l2lw == 8` and so `width`
starts at 9 bits. Printing the bits on byte boundaries give:
offset xoffset ASCII hex binary
000001 0x0001 . 0x00 0b_0000_0000
@@ -452,8 +452,8 @@ bytes and then an 0xD9.
## Actual Pixel Indexes
It is also not an official format, but for reference, the
`test/testdata/bricks-nodither.indexes` contains the decoded
`test/testdata/bricks-nodither.indexes.giflzw` data:
`test/data/bricks-nodither.indexes` contains the decoded
`test/data/bricks-nodither.indexes.giflzw` data:
offset xoffset ASCII hex binary
000000 0x0000 . 0xDC 0b_1101_1100
@@ -486,6 +486,6 @@ We have 19200 = 160 × 120 pixels. The top row starts with five pixels with RGB
values indexed by 0xDC (or 220 in decimal, and recall that the GCT maps color
220 to {0x0F, 0xA7, 0xF6}), the bottom row ends with index 0x4F.
Indeed, opening `test/testdata/bricks-nodither.gif` in an image editor should
Indeed, opening `test/data/bricks-nodither.gif` in an image editor should
verify that the top left pixel's RGB is {0x0F, 0xA7, 0xF6}, and likewise the
bottom right pixel's RGB is {0x01, 0x24, 0x63}.