Seven characters in 5x8.flf should have height 9

[ronnodas]

This is sort of related to #69 and #123 which fixed the header to say the height is 9 since that was true for most of the characters. But the last 7 of the 102 required characters had height 8 (as in the old incorrect header). These are all empty (as in have no subcharacters) and outside the ascii range (as in their characters codes) which is probably why this didn't show up in tests.

[peterbrittain]

Looks plausible to me. Probably should run a quick test... Do the other characters after this section display correctly now?

[ronnodas]

I believe so from manual testing. I didn't figure out where/how the automated tests go.

[peterbrittain]

You can just run tests.py for the test suite (assuming you also have figlet installed). They won't test this case because not all fonts have control codes. In fact, after trying to add a test, I discovered figlet deliberately drops them!

[pwaller]

Please include what you did for manual testing in the body -- that makes it easier for me to understand what's happening.

[ronnodas]

Please include what you did for manual testing in the body -- that makes it easier for me to understand what's happening.

Sorry, I misread the question. I had just checked that this change didn't mess up standard text.

For the characters after the change, it looks like the character codes don't really match the glyphs either, eg the ones tagged 26 and 27 should probably be 0x2264 and 0x2265 respectively.

[peterbrittain]

Yeah - I was trying to figure out the logic for the assignments, but these don't match anything I've seen in the past (e.g. codepage 437).

I've also discovered that pyfiglet cannot display them anyway - the current logic to load extended characters requires a hex index. Which is a happy coincidence, given that figlet ignores all control codes.

Suggest we delete these extended chars as they would never have been displayed by figlet.

[ronnodas]

Here are my best guesses for what these characters are supposed to be:

1 C001 -> 0x2666, possibly others
2 C002 -> ???
3 C003 -> 0x09
4 C004 -> 0x0C
5 C005 -> 0x0D
6 C006 -> 0x0A
7 C007 -> 0x25CB, many other choices
8 C010 -> 0x00B1
9 C011 -> 0x0A again?
10 C012 -> 0x0B
11 C013 through 25 C031 -> some subset of the box drawing block, 0x2500 -- 0x257F, but some choices here
26 C032 -> 0x2264
27 C033 -> 0x2265
28 C034 -> 0x03C0
29 C035 -> 0x2260
30 C036 -> 0x00A3
31 C037 -> 0x00B7, possibly others
127 C17 -> ???
0 C000 -> ???

Does pyfiglet handle the "0 = missing character" rule? If so, keeping the 0 might make sense.

[peterbrittain]

Nope. Pyfiglet will ignore missing characters instead.