Teaser peek at Version 6 XML Format

Thank you, Jorge.  It looks like you're off to a good start, however, I'd like to echo some comments I just made in the thread Phil Torrrone started back in October:

On 14.09.2011 12:30, Brian Lalor wrote:

...

First is the use of the word "ASCII", which I'm just hoping is sloppy technical jargon for "text".  ASCII is a very limited character encoding that has been surpassed by UTF-8, which is able to represent characters like "é" and "€" (although, dollars to donuts that Jive screws those up in this reply, too).

It's UTF-8, of course.

I'm also bothered by the lack of indenting in the screenshot above, as it doesn't show much attention to detail.

We did this on purpose to keep the file size as small as possible.

What's most important is that a robust, stable, versioned XML schema is published, and that the tools that CadSoft provide for parsing and generating XML documents strictly adheres to it and provides meaningful errors when validation fails.  Do not fall into the trap of building a loosely-defined schema and then throwing errors in the program logic when the attribute you defined as an xsd:string isn't a valid floating-point number.  A good schema will allow you to use tools like xmllint to make sure the XML you're generating from gEDA or Kicad (or from the bash script you wrote for generating your schematics before GUI tools became cool) is valid and stands a solid chance of being usable by EAGLE.

We'll do our best - but in the end, EAGLE itself is the judge of what's

correct

Klaus Schmidinger

--

_______________________________________________________________

Klaus Schmidinger                       Phone: +49-8635-6989-10

CadSoft Computer GmbH                   Fax:   +49-8635-6989-40

Pleidolfweg 15                          Email:   kls@cadsoft.de

D-84568 Pleiskirchen, Germany           URL:     www.cadsoft.de

_______________________________________________________________

Am 14.09.2011 18:04, schrieb Klaus Schmidinger:

It's UTF-8, of course.

EAGLE's user interface already supports Unicode character ENTRY since

some time. Unfortunately, special Unicode characters are just NOT taken

from the user interface to the program itself, but are either converted

to question marks or to nothing.

The storage format for EAGLE 6 will be UTF-8. Wonderful. Therefore, both

user interface AND storage will be able to cope with non-ASCII (above

#127) characters. Will perhaps even the VECTOR font get all additional

characters to support the Unicode ranges "C0 controls and basic latin"

(that's ASCII 32..127) PLUS "C1 controls and Latin-1 supplement" (lots

of stuff from ANSI 128..255)? That would at least make all western

european people happy...

And, since several people already complained about the vector font not

really LOOKING nice: Good old Mr. Hershey didn't only design the basic

characters of EAGLE's vector font, but also a highly-readable stroked

font that looks very nice and is sometimes called "Simplex" (yes, that

stuff came with Borland compilers in EAGLE's Pascal times), and which

even has Greek characters in (Borland's Greeks didn't look convincing).

The Hershey fonts should be license-free and easy to patch to include

accents etc. For font samples and data, take a look at

   http://paulbourke.net/dataformats/hershey/

(some characters would need to be reduced to single-stroked ones).

You will even find a Japanese Hershey set available there. Hiragana and

Katakana are supplied, but I'm not so sure whether the rest is really

useful (not knowing Japanese at all), or whether it's really only a

fraction of what's necessary.

Andreas Weidner

Am 14.09.2011 18:04, schrieb Klaus Schmidinger:

It's UTF-8, of course.

EAGLE's user interface already supports Unicode character ENTRY since

some time. Unfortunately, special Unicode characters are just NOT taken

from the user interface to the program itself, but are either converted

to question marks or to nothing.

The storage format for EAGLE 6 will be UTF-8. Wonderful. Therefore, both

user interface AND storage will be able to cope with non-ASCII (above

#127) characters. Will perhaps even the VECTOR font get all additional

characters to support the Unicode ranges "C0 controls and basic latin"

(that's ASCII 32..127) PLUS "C1 controls and Latin-1 supplement" (lots

of stuff from ANSI 128..255)? That would at least make all western

european people happy...

And, since several people already complained about the vector font not

really LOOKING nice: Good old Mr. Hershey didn't only design the basic

characters of EAGLE's vector font, but also a highly-readable stroked

font that looks very nice and is sometimes called "Simplex" (yes, that

stuff came with Borland compilers in EAGLE's Pascal times), and which

even has Greek characters in (Borland's Greeks didn't look convincing).

The Hershey fonts should be license-free and easy to patch to include

accents etc. For font samples and data, take a look at

   http://paulbourke.net/dataformats/hershey/

(some characters would need to be reduced to single-stroked ones).

You will even find a Japanese Hershey set available there. Hiragana and

Katakana are supplied, but I'm not so sure whether the rest is really

useful (not knowing Japanese at all), or whether it's really only a

fraction of what's necessary.

Andreas Weidner