feat: add isGurmukhi function to check if string is Unicode Gurmukhi

[sarabveer]

Closes #140

[sarabveer]

Need to add tests

[Harjot1Singh]

I'd suggest calling it isUnicodeGurmukhi or isGurmukhiUnicode, since isGurmukhi can be easily confused with "Is this string Gurmukhi" btw.

[sarabveer]

I'd suggest calling it isUnicodeGurmukhi or isGurmukhiUnicode, since isGurmukhi can be easily confused with "Is this string Gurmukhi" btw.

ASCII is ASCII, it isn't Gurmukhi. There isn't a way to test if a string is from an ASCII Gurmukhi font, and if there was, English would also return true for such a function.

[coveralls]

Pull Request Test Coverage Report for Build 301

• 8 of 8 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-0.8%) to 99.2%

---

Totals
Change from base Build 297:	-0.8%
Covered Lines:	95
Relevant Lines:	95

---

💛 - Coveralls

[bhajneet]

Since this PR is closing #140, perhaps we can also fix toAscii function in this PR?

[sarabveer]

Since this PR is closing #140, perhaps we can also fix toAscii function in this PR?

@Harjot1Singh how would this be done?

[Harjot1Singh]

Since this PR is closing #140, perhaps we can also fix toAscii function in this PR?

@Harjot1Singh how would this be done?

toAscii = (text) => isAscii(text) ? text : otherStuffInFunction

Something like that. I'm not sure if we'd want exhaustive enabled or not.

[bhajneet]

Since this PR is closing #140, perhaps we can also fix toAscii function in this PR?

@Harjot1Singh how would this be done?

toAscii = (text) => isAscii(text) ? text : otherStuffInFunction

Something like that. I'm not sure if we'd want exhaustive enabled or not.

Wouldn't it be better to fix toAscii from not running the post rules block if it's already in ascii?

[Harjot1Singh]

Since this PR is closing #140, perhaps we can also fix toAscii function in this PR?

@Harjot1Singh how would this be done?

toAscii = (text) => isAscii(text) ? text : otherStuffInFunction
Something like that. I'm not sure if we'd want exhaustive enabled or not.

Wouldn't it be better to fix toAscii from not running the post rules block if it's already in ascii?

Wouldn't this mean running some of toAscii, followed by then running isAscii? Why do extra work?

[sarabveer]

There is no isAscii it is isGurmukhi

[Harjot1Singh]

There is no isAscii it is isGurmukhi

My bad, isGurmukhi instead of isAscii

[bhajneet]

Wouldn't this mean running some of toAscii, followed by then running isAscii? Why do extra work?

Sorry, I misread. Using isExtendedAscii/isGurmukhi would be fine in the line you said. However since isGurmukhi's exhaustive parameter requires all characters to be in unicode to pass, I do not believe it will work for shabados/desktop. (In the sense that you are trying to convert something toAscii on each keystroke of unicode keyboard and each keystroke is supposed to be converted toAscii).

[bhajneet]

If you use "some" instead of "every" in isGurmukhi function then it isn't true to the name of the function. Maybe better to check "every"/exhaustive for extended ascii chars and if false then converting to ascii?

Edit: or rename func to hasGurmukhi

[sarabveer]

However since isGurmukhi's exhaustive parameter requires all characters to be in unicode to pass, I do not believe it will work for shabados/desktop.

This is why I believe and argue toAscii should not be "fixed". toAscii is doing its job, but due to the nature of ASCII, the replacements have to be done after and cannot be done before.

isGurmukhi should be used by implementations of toAscii and check before calling the toAscii function.

[bhajneet]

This is why I believe and argue toAscii should not be "fixed". toAscii is doing its job, but due to the nature of ASCII, the replacements have to be done after and cannot be done before.

isGurmukhi should be used by implementations of toAscii and check before calling the toAscii function.

That would still fail my use case.

[sarabveer]

This is why I believe and argue toAscii should not be "fixed". toAscii is doing its job, but due to the nature of ASCII, the replacements have to be done after and cannot be done before.
isGurmukhi should be used by implementations of toAscii and check before calling the toAscii function.

That would still fail my use case.

How so? Every keystroke can be run through isGurmukhi without exhaustive. This issue would be out of scope for gurmukhi-utils.

[bhajneet]

How does this work for us if we're converting each keystroke input for search query?

[sarabveer]

Why is this issue being raised now? The functionality you have shown is expected.

Input is supposed to be explicit, either Gurmukhi or ASCII. We agreed mixing would not be allowed.

EDIT: This is reflected in the tests as well https://github.com/ShabadOS/gurmukhi-utils/blob/isGurmukhi/test/isGurmukhi.spec.js#L13

[Harjot1Singh]

Add an example to examples.js too, as well as the example section at the top of README.hbs, then good to go