Now support web apps

[NanoMichael]

Now we support web apps, check this demo out.

And I'm going to release version 1.0.0, but before that, we have a lot of non-trivial work to do.

And most important, I'm planning to rename the project to TinyTeX, as its main goal is: fast and tiny. But I'm wondering if is it appropriate. Can you guys give me some advice? @sp1ritCS @jeremysanders @blackhole89 @PikachuHy 👀

[PikachuHy]

good job.
I will review code.

[PikachuHy]

• Do we need to rename all LaTeX to TinyTex?

https://github.com/NanoMichael/cLaTeXMath/blob/dc32540664276a00805b21a0fb5cd6418bacdb54/CMakeLists.txt#L2

https://github.com/NanoMichael/cLaTeXMath/blob/ae70d932fb78e45e2c8a8e029ee95fe52820ee77/platform/qt/CMakeLists.txt#L3

• Confusing although compiled.

https://github.com/NanoMichael/cLaTeXMath/blob/ae70d932fb78e45e2c8a8e029ee95fe52820ee77/lib/otf/clm.cpp#L24
What does this code mean? convert (*(p + i)) to T or convert (*(p + i)) << ((shift - i) * 8) to T?

comiled with warning

C:\code\cLaTeXMath\lib\otf\clm.cpp(24): warning C4805: '|=': unsafe mix of type 'T' and type 'int' in operation
        with
        [
            T=bool
        ]
C:\code\cLaTeXMath\lib\otf\clm.cpp(89): note: see reference to function template instantiation 'T tex::BinaryReader::read<bool>(void)' being compiled
        with
        [
            T=bool
        ]
C:\code\cLaTeXMath\lib\otf\clm.cpp(24): warning C4805: '|': unsafe mix of type 'bool' and type 'int' in operation

• Add conan support

public our library to conan-center: https://conan.io/center/

Conan is an open source, decentralized and multi-platform package manager to create and share all native binaries.

• suggestion: Add prebuilt clm files to v1.0.0 release package. (e.g. XITSMath-Regular.otf)
• suggestion: Use a separate GitHub organization. (e.g. tinytex)

[LyesSaadi]

please don't, @LyesSaadi and me would like to avoid any licensing issues that would come with that. Instead we'll pull the otf files from their respective packages (i.e. texlive-xits-fonts) and build the clm file during buildtime.

oh no, halp, no licensing issues. please.

[PikachuHy]

conan just provides a "linux" build

conan is written by python.
It is just a package manager like vcpkg, which we used in ci.

from conan doc

Conan is a dependency and package manager for C and C++ languages. It is free and open-source, works in all platforms ( Windows, Linux, OSX, FreeBSD, Solaris, etc.), and can be used to develop for all targets including embedded, mobile (iOS, Android), and bare metal. It also integrates with all build systems like CMake, Visual Studio (MSBuild), Makefiles, SCons, etc., including proprietary ones.

maybe we can also add brew support (for mac user).

[sp1ritCS]

conan is written by python.
It is just a package manager like vcpkg, which we used in ci.

Yeah, but that's not what I ment. Besides "ALT Linux", there is no Linux distro which officially offers conan (https://pkgs.org/search/?q=conan). That leads me to asume that it's not that popular (even more so the fact, as they do not even ship a gtkmm package there).

Additionally, the problem is, that they do not offer builds for various distros, but only one "Linux". But obv. different Linuxes may ship with different versions of libraries and thus there might be symbols that will be missing on some distros.

maybe we can also add brew support (for mac user).

see: https://github.com/sp1ritCS/homebrew-tap/blob/master/Formula/clatexmath.rb

[PikachuHy]

there is no Linux distro which officially offers conan

yes

conan is installed by pip install conan

just provide a option to our user.
native package manager(apt or yum) or 3rd package(vcpkg, conan, etc) is ok.

[sp1ritCS]

conan is installed by pip install conan

yes, but that is obv. not officially offered. instead if pulls the conan source of pypi which is not validated by any of the distros.

if it was popular, there'd be some kind of (python38-)conan package in the official repos, like pip itself is (python38-pip).

I'm just considering if the effort of publishing it as a package on a platform that nobody uses 😦, but if I want to I don't think anybody will stop you :).

[PikachuHy]

I met a error.

see #97

[NanoMichael]

Seems that the version of the fontforge is too low? By the way, now we can find the prebuilt clm resources from here.

[PikachuHy]

only otf files here

[NanoMichael]

I'll check it soon

[NanoMichael]

only otf files here

fixed by 2bf07ba

[LyesSaadi]

:O, the web app demo looks so cool (and works quite well)! I guess it's through WASM? (just checked, it seems so)

The TinyTex rename doesn't have much impact on packaging for Fedora since nothing was pushed yet (and, anyway, renames are not hard to handle). Also, the new name seems to be more discoverable (I'd image someone would just type tiny latex library or tiny tex library to find it).

[NanoMichael]

Well, I've found a project (see here) already named TinyTeX 😂... there are some choices:

• FastTeX, sounds a little bit silly
• MathTeX or TeXMath, not bad
• TinyMath, already have one, see here

Any suggestions?

[NanoMichael]

yeah, it is too hard for those non-Greek users to input the 'µ', but 'µTeX' as the logo is great. Do we consider using meutex? 🤔

[sp1ritCS]

I have no idea why you'd call it meutex. I'd say name it microtex, given that µ is the SI prefix for micro (10^{-6}).

Also I think all ANSI keyboards have the ability to input µ with AltGr+M (I think), but yeah might be hard for Americans :)

[NanoMichael]

I have no idea why you'd call it meutex.

It is a homonym for 'mu' 😂 I think microtex is good. Please see here, I got your point.

[sp1ritCS]

It is a homonym for 'mu' 😂

huh, then maybe pronunciation varies in different languages/cultures. In Germany we say mü (IPA [myː]) not [mɔʏ̯] or [meuː]

[NanoMichael]

huh, then maybe pronunciation varies in different languages/cultures. In Germany we say mü (IPA [myː]) not [mɔʏ̯] or [meuː]

Oh, I didn't realize it at all.

So, the answer becomes clear, in conclusion, use 'microtex' as namespace, and \mu\TeX for the in-renderer name. I'll make a PR for renaming.

[PikachuHy]

And I'm going to release version 1.0.0

Some problems I found when studying or reviewing this project.

[PikachuHy]

suggestion: Add unit test

[PikachuHy]

• typo

MicroTeX/lib/atom/atom_char.h

Line 20 in 9cae240

* Row will mark certain CharSymbol atoms as a text symbol. Subsup wil use

[PikachuHy]

• Logical expression is always true

MicroTeX/lib/core/parser.cpp

Line 719 in 9cae240

if (ch != ' ' || ch != '\t' || ch != '\r') break;

[PikachuHy]

• unnecessary L

After using std::string, L is redundant.

MicroTeX/lib/core/parser.cpp

Line 266 in 9cae240

char ch = L'\0';

MicroTeX/lib/core/parser.cpp

Line 279 in 9cae240

if (ch == L'\0') return "";

[PikachuHy]

• refactor: use isAlpha in utils.h

MicroTeX/lib/core/parser.cpp

Line 270 in 9cae240

if ((ch < 'a' || ch > 'z')

[PikachuHy]

• typo

everywhere???

MicroTeX/lib/atom/atom_row.h

Line 140 in 9cae240

static bool _breakEveywhere;