Get unexpected result

[ngekoding]

I just follow the instructions, but got the wrong result.

Here is my code to extract the data

const tajweedRules = require("./tajweed.hafs.uthmani-pause-sajdah.json");

const rules = tajweedRules[8].annotations; // 2:2

const text = "ذَٰلِكَ ٱلْكِتَٰبُ لَا رَيْبَ ۛ فِيهِ ۛ هُدًى لِّلْمُتَّقِينَ";

rules.forEach(rule => {
  const subText = text.substring(rule.start, rule.end);
  console.log(rule.start + " - " + rule.end + ": " + rule.rule);
  console.log(subText);
});

And here is the result

Am I doing something wrong?

Thanks in advance

[cpfair]

It appears that tanzil.net has made some minor encoding changes to the text, so the indices no longer align. I have attached the original quran-uthmani.txt file I used when when generating the rule files - in shaa Allah I will look at regenerating them to match the current tanzil.net format.

quran-uthmani.txt (downloaded from Tanzil.net ca. Apr 6, 2017).

[ngekoding]

Jazaakallahu khairan, I will try to use it.

[ngekoding]

Tested with 2:2 with no problem, but with 2:3 the problem still exists.

The code still same with the above, only for the start changed as because the substring index begin from 0;

rules.forEach(rule => {
  const subText = text.substring(rule.start - 1, rule.end); // changed to: rule.start - 1
  console.log(rule.start + " - " + rule.end + ": " + rule.rule);
  console.log(subText);
});

any idea?

[ngekoding]

If I use original code (without - 1 for the start), it worked for 2:3, but 2:2 got problem.

[cpfair]

Based on the reference material used in generating the rulesets, the dal is not included in that rule. E.g. http://recitequran.com/2:2 (highlights the fathatain on the dal, but not the dal itself). And, due to the way the rules are represented in this system, they can apply only to a contiguous span of codepoints (whereas in this case it may be better if it could cover two disjoint spans, as seen in the ReciteQuran link above).

[ngekoding]

Thank you, so this is my misunderstanding about the implementation of the rules.

[korbav]

It appears that tanzil.net has made some minor encoding changes to the text, so the indices no longer align. I have attached the original quran-uthmani.txt file I used when when generating the rule files - in shaa Allah I will look at regenerating them to match the current tanzil.net format.

quran-uthmani.txt (downloaded from Tanzil.net ca. Apr 6, 2017).

Wonderful. That could also explain the issues I raised in another comment.

I just follow the instructions, but got the wrong result.

Here is my code to extract the data

const tajweedRules = require("./tajweed.hafs.uthmani-pause-sajdah.json");

const rules = tajweedRules[8].annotations; // 2:2

const text = "ذَٰلِكَ ٱلْكِتَٰبُ لَا رَيْبَ ۛ فِيهِ ۛ هُدًى لِّلْمُتَّقِينَ";

rules.forEach(rule => {
  const subText = text.substring(rule.start, rule.end);
  console.log(rule.start + " - " + rule.end + ": " + rule.rule);
  console.log(subText);
});

And here is the result

Am I doing something wrong?

Thanks in advance

I noticed you're using Javascript as well.
It seems that you really can get it working naively.
As suggested by another user it seems obligatory to take care of the sleeping marks offset to have it more or less working.
Even with that though, and maybe because the uthmani script got updated, I was unable to get the final result 100% aligned with the original books.

[ngekoding]

@korbav Currently I use the tajweed version by Quran.com API. I really don’t dare to take any risks for use using this way for now.

[korbav]

Quran.com API

That seems to be a great option.
This project is awesome too, and I'm sure we're really close to get a great alternative to the Quran.com API with it!

[cpfair]

That API looks nice. I've added it to the readme as a suggested alternative, given that this project hasn't been maintained in some time.