Update on Press (the PDF compiler). I haven’t worked on it at all lately, but I wanted to document its current state for history’s sake, and as part of working in public. (I’ve also been sitting on this post for over a year.)
Back in 2017 I did end up re-architecting Press to use Low Ink as an intermediate page description language. In the process, Low Ink changed from a JSON-based idea to this:
:bleedbox x=0.125in y=0.125in w=5.75in h=8.75in
:fontmap family=helv weight=regular style=normal standard=Helvetica
:translate x=72 y=72
:translate x=0 y=1040
:line x1=0 y1=0 x2=1080 y2=0
:font family=helv size=14pt
:text x=1085 y=-3 text="ascender"
# filled glyph
:translate x=1320 y=240
:moveto x=0 y=0
:pathto x=400 y=300 cx1=120 cy1=300 cx2=140 cy2=300
:pathto x=320 y=200 cx1=540 cy1=300 cx2=320 cy2=180
:lineto x=350 y=350
:lineto x=450 y=250
:lineto x=150 y=0
:moveto x=200 y=200
:lineto x=200 y=250
:lineto x=250 y=250
:lineto x=250 y=200
:lineto x=200 y=200
It was intended to be a fairly low-level wrapper on the PDF format, with the idea being that other libraries/apps would provide more ergonomic abstractions on top of it.
I initially used Python because Press started out as a library, but with the pivot to a compiler model, I think Go or Rust would probably end up being a better choice (Rust would make integrating HarfBuzz a bit easier, at any rate).
To my 2021 eyes, the language design isn’t particularly elegant. I like that the parameters are named (clarity), but for most of the commands there aren’t actually that many parameters, because many of the settings that would normally be parameters are separate commands. For parameters that are clearly unambiguous, the names hamper readability. For example, I think something like this might be better:
:line 0,0 to 1080,0
I’ve also thought that
pop could potentially be clearer as curly braces, and that the initial colons aren’t really necessary:
line 0,0 to 1080,0
font 14pt helv
text 1085,-3 "ascender"
My initial reason for building Press was to have an easy, programmable cross-platform way to create language chart PDFs (so I could move away from PlotDevice/DrawBot), and what I’ve realized (acknowledging that I haven’t really been making language charts in recent years) is that there are some other, better options now.
One that seems decent is SVG, converted to PDF by way of Inkscape. Initial tests here seem like it would probably work fine.
Another promising option that I admittedly haven’t looked into very much yet is Paged.js. HTML and CSS are already great for declarative typesetting, and the more I’ve thought about programmatic typesetting, the more this model seems to be the future I’d want to work with (and not just because of parity with web, though that makes it much more compelling).
tl;dr I don’t see myself continuing on with Press, so we may as well call a mortem on it.
Blogging is low on the priority list at the moment, thanks to school. The preliminary classes for the master’s degree are going well. I’m writing assembly for my computer systems class, and I have to say, I really like assembly. (No sarcasm.) It’s beautiful and simple in a way I didn’t expect. I don’t see myself using it much, but it’s a good tool for the belt.
Oh, with Press, I realized a few days ago that it’s a good candidate for the first implementation of Low Ink (a JSON-based page description language that compiles to PDF). I’ll be re-architecting that part of Press so that it uses Low Ink. Also hoping to finish up the text part of Press (HarfBuzz, etc.) soon so that it’s usable for more than just basic drawing. (I’m dealing with font subsetting and encoding stuff at the moment.)
After a break of several months, I’m getting back to working on Press. Status is pretty much the same as last time I posted about it. (It’s actually even a little more behind than that, since I had HarfBuzz Python bindings working then, but now — after upgrading to macOS Sierra — I’m running into issues with PyGObject’s introspection module. I may end up having to write my own HarfBuzz bindings with CFFI. We’ll see.)
The high-level roadmap right now: get font embedding to work correctly, add support for embedding images (which should be fairly easy, I think), integrate ICU for language analysis and HarfBuzz for shaping, and add color space support.
As of now, I plan to use Press for making language charts (which I’ve been using PlotDevice for) and picture books. Once it’s to the point where I can do that, then I’ll start on Ink (low-level typesetting engine, intended for typesetting books, and higher-level rule-based engine for making it easier to work with).
I’ve been hand-coding PDFs in Vim, reading the PDF spec to learn how things work. It’s fascinating. My first, extremely simple PDF:
1 0 obj << /Type /Catalog /Pages 2 0 R >>
2 0 obj << /Type /Pages /Kids [3 0 R] /Count 1 >>
3 0 obj << /Type /Page /Parent 2 0 R /Resources 4 0 R /MediaBox [0 0 500 800] /Contents 6 0 R >>
4 0 obj << /Font << /F1 5 0 R >> >>
5 0 obj << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>
6 0 obj
<< /Length 44 >>
BT /F1 24 Tf 175 720 Td (Hello World!) Tj ET
0000000000 65535 f
0000000010 00000 n
0000000059 00000 n
0000000116 00000 n
0000000220 00000 n
0000000263 00000 n
0000000333 00000 n
trailer << /Size 7 /Root 1 0 R >>
It’s not as bad as it looks, I promise. (I’m doing PDF 1.4 because CreateSpace doesn’t seem to support higher versions of the spec.)
Anyway, I’ve been reading through chapter 5 of the spec, learning how text works in PDF. I’ve learned how to modify character spacing with
Tc, word spacing with
Tw, leading with
TL, and individual glyph positions with
TJ (not sure yet if I can change vertical positioning or not). I’ve also learned how to change the text color. It’s all been fairly straightforward.
As part of this, I’ve used Hex Fiend (an OS X hex editor) to pry apart some simple PDFs I made with PlotDevice, to see how things were encoded. The streams themselves are generally compressed through Flate compression (opposite of deflate, har har), and I found this script to easily decode the streams:
input = sys.argv
output = sys.argv
with open(input, 'rb') as f:
buffer = f.read()
decomp = zlib.decompress(buffer)
with open(output, 'w') as f:
I copied each stream in hex from Hex Fiend, pasted it into a file, ran the Python script on it, and it would output decoded text to a new file.
Things I don’t know/understand yet, which are legion:
- How to encode Unicode (I’m not to this point of the spec yet, but I believe it involves CID fonts and using cmaps to map glyph codes or something like that).
- How to take a font name and, in a cross-platform way, get the path to the font file so I can embed it and also use it with HarfBuzz.
- How to take the output of HarfBuzz (a list of glyphs with position coordinates for each) and use that in positioning the glyphs in the PDF. I believe HarfBuzz will handle parsing the OpenType features of the font, but I’m not positive on that. I did get HarfBuzz Python bindings working, though, and I plan to play around with it soon.
- Whether I need to use FreeType at all. I might need it for font metrics, but HarfBuzz might give me everything I need there.
- When typesetting multiple lines, I don’t know whether it’s best to use the PDF built-in support (
TL and such), or to set each line manually as its own text object. The built-in support seems better, though I don’t know if that limits what’s possible.
At some point soon — I think when I start embedding fonts — doing this by hand in Vim will stop being as feasible, and at that point I’ll start writing Python to manage the PDF creation process for me. For now, though, it’s easier to just edit the PDF manually.