Font Forum

Hello,

I am hoping to filter my font preview to only show ones that include ALL characters in the Preview text area (in other words, fonts that aren't red).

This Vietnamese text would ideally only return a few results:

Trân trọng báo tin Lễ Thành hôn và Vu Quy của con chúng tôi

Although this would be a nice feature, it is unfeasible right now.

Ardy Haze wrote:This Vietnamese text would ideally only return a few results:

Trân trọng báo tin Lễ Thành hôn và Vu Quy của con chúng tôi

If (as I do) you only have a few font families that aren't red with that text sample, it shouldn't take long to tag them as Vietnamese and/or add them to a group so that you can filter them easily.

Erwin Denissen wrote:Although this would be a nice feature, it is unfeasible right now.

Would you mind explaining why it's not feasible? Seems like if you can get text to change red, you can get it to disappear too.

Alfred wrote:

Ardy Haze wrote:This Vietnamese text would ideally only return a few results:

Trân trọng báo tin Lễ Thành hôn và Vu Quy của con chúng tôi

If (as I do) you only have a few font families that aren't red with that text sample, it shouldn't take long to tag them as Vietnamese and/or add them to a group so that you can filter them easily.

I appreciate the suggestion, but this is only one case example where I would find this feature useful. I'd also like it for ampersands and other special characters.

We only know if characters are available when we preview the font. Previewing thousand fonts at one time would take minutes. People with 50000 fonts would have to wait for hours.

Erwin Denissen wrote: ↑Thu Jun 09, 2016 9:14 pm People with 50000 fonts would have to wait for hours.

With MainType 8, 64-bit, it only takes about a minute for my PC to build a list of 75,000 fonts, and scrolling through it has font previews almost as fast as it can scroll. The real delay is rendering the scroll bar and the characters on-screen. An in-memory creation of the preview would be much faster.

Also, a lot of this sort of checking can be done in the background, while the user isn't doing anything, so it would be available instantly as far as they are concerned. Even if it took overnight to run, the information can be stored in the database for very fast access.

In the listed example of "does the preview have all characters available?", although it can't be stored in the database (unless you store it only for each saved preview text, and then update the database when the user creates a new saved preview text), it actually doesn't require any display to figure out, as you have the font in the vault, so you know exactly what characters are available, and you know what the preview text consists of. The list can then be filtered based on that comparison.

There are a lot of other data that might take hours to generate, but could be stored in the database, like the answers to the questions that "Identifont" uses to figure out what font you are looking at. This could then be used to help the user find similar fonts without requiring the font to have correct PANOSE (or any other metadata), and no tagging on the part of the user.

This could come in handy with many languages, even Middle European ones (find the fonts containing ij or ß ...).

Regarding unfeasible, in should be possible to compact glyph subsets of many fonts to ranges (has 0x00-0xff, 0x4000-0x40ff, 0x20ac-0x20ac,...) and applying a plane sweep algorithm.

+1

Need this - or a search tool - to find fonts for minority languages.

We are in 2024 now, and I still need a better way to search and filter (by certain glyphs contained in fonts). And to store such a filter and use it any time.

And best would be to be able to apply any filter on top of any group.

So groups could be for content-features, like "suitable for a certain client".

Filters could be for technical-features, like "font contains German capital ß, i.e. ẞ".




Or if the creators prefer groups-only:
We would need a way to search or filter fonts and turn them into a group efficiently with a click.

And to make AND combinations from two or more groups. This would give users a way to use only fonts which fulfil several needed criteria.

Or we could dream and have syntax: Combine several groups with AND, OR, NOT...