Font Generated code changed when build the project without change Font.

Question

Hi EW teams,

The following situations often occur without changing the Font configuration, but after build, the file undergoes significant changes. This causes confusion for me, as I am unsure which of the two files is correct.

I would like to know what the last parameter specifically refers to. Is it the save location?          -> EW_GLYPH( 0x0031, 1, -9, 5, 9, 7, 0x000005DF ),               /* '1' */

Is this kind of change normal, and can it be ignored during version change management?

 

 

Best Regards，

Chris Zhao 

Answer 1

Hello Chris,

I would like to know what the last parameter specifically refers to. Is it the save location?          -> EW_GLYPH( 0x0031, 1, -9, 5, 9, 7, 0x000005DF ),               /* '1' */

yes, this is an offset within the memory area containing the compressed pixel data.

Is this kind of change normal, and can it be ignored during version change management?

It is not normal. If there is no modification in the font configuration and you are using the exact the same font file, the outputs should be equal.

Question: Do you generate the code on different PCs? In such case, ensure that on both machines exact the same version of the True Type font is available/installed.

Also, please check the Log window for eventual warnings related to the font conversion.

I hope it helps you further.

Best regards

Paul Banach

Comments

Hello Paul,

yes, this is an offset within the memory area containing the compressed pixel data.

Does it mean that the changes of memory area  will not affect the display of text. It seems that the previous parameters have not changed. I roughly checked the display of the relevant fonts, but I couldn't find any display errors.

Question: Do you generate the code on different PCs? In such case, ensure that on both machines exact the same version of the True Type font is available/installed.

Also, please check the Log window for eventual warnings related to the font conversion.

No, It is the same PC . And there is no warning in the log window.

 

Best Regards

Chris Zhao 

 

 

---

Hello Chris,

Does it mean that the changes of memory area  will not affect the display of text. It seems that the previous parameters have not changed. I roughly checked the display of the relevant fonts, but I couldn't find any display errors.

The changes should not have any impact on the text display. It seams the font data is different.

No, It is the same PC . And there is no warning in the log window.

Ok. Under these circumstances, I would expect the same results.

Are there any particularities on your PC? E.g. are you switching Windows language between different languages?

Have you eventually observed the difference after updating Windows? May be new font version have been installed by Windows?

If you could narrow down the conditions leading to the different code, it would help us to understand the cause.

Best regards

Paul Banach

---

Hello Paul，

Are there any particularities on your PC? E.g. are you switching Windows language between different languages?

It seems that this is the reason, because some software supports languages restrictions, I have the action of switching Windows languages. I just tried it and indeed reproduced this problem. 

So I have to ensure that the Windows language is fixed  every time I build the project, right?

Best Regards

Chris Zhao 

---

Hello Chris,

So I have to ensure that the Windows language is fixed  every time I build the project, right?

Very interesting observation. Apparently, the appearance of a glyph changes depending on the selected Windows language. This may occur when a font file contains language specific variants of glyphs.

According to the screenshot from your original post, the affected glyph is % (percent sign) - this is because all glyphs following it have different position in the compressed data stream:

So I have to ensure that the Windows language is fixed  every time I build the project, right?

Not necessarily. If you tested the GUI and you haven't detected any differences in text output it could be acceptable to ignore the different code. Also, you could test the appearance of the glyph % with different languages. May be the difference is subtil and can be ignored?

However, if you expect exactly the same code, then yes would would need to think about the right language before generating code.

Best regards

Paul Banach