<![CDATA[Curl Global Community

<![CDATA[Curl Global Community - Robert blog]]> https://communities.curl.com/ Sun, 12 Apr 2026 20:21:27 +0000 MyBB <![CDATA[Non-selectable characters amidst text ]]> https://communities.curl.com/showthread.php?tid=1108 Mon, 08 Sep 2014 18:09:05 +0000 https://communities.curl.com/showthread.php?tid=1108 [Image: non-selectable_chars_sqr_750.png]

[Image: non-selectable_chars_sqr_750.png]

Here is the data for the first rows of displayed :

Code:
{krow "0014","0000",'〇'}

{krow "0002","2850",'一'}

{krow "0003","1688",'ニ'}

{krow "0437","2851",{on-yomi-only {CharGraphic '丁'}}}

{krow "0009","2854",{CharGraphic '七'}}

{krow "0018","2542",{on-yomi-only 万}}

{krow "0657","2885",'丈'}

{krow "0004","1689",'三'}

{krow "0041","2876",'上'}

{krow "0040","2862",'下'}

{krow "0049","2890",{on-yomi-only 不}}

{krow "0858","2887",'与'}

As you can see, two formatting fields have CharGraphic and not merely a char.
Yet this is not visible to the user, something that I am finding useful.

]]> [Image: non-selectable_chars_sqr_750.png]

Here is the data for the first rows of displayed :

Code:
{krow "0014","0000",'〇'}

{krow "0002","2850",'一'}

{krow "0003","1688",'ニ'}

{krow "0437","2851",{on-yomi-only {CharGraphic '丁'}}}

{krow "0009","2854",{CharGraphic '七'}}

{krow "0018","2542",{on-yomi-only 万}}

{krow "0657","2885",'丈'}

{krow "0004","1689",'三'}

{krow "0041","2876",'上'}

{krow "0040","2862",'下'}

{krow "0049","2890",{on-yomi-only 不}}

{krow "0858","2887",'与'}

As you can see, two formatting fields have CharGraphic and not merely a char.
Yet this is not visible to the user, something that I am finding useful.

]]> <![CDATA[Curl with HTML widget + CintaNotes]]> https://communities.curl.com/showthread.php?tid=1107 Sun, 07 Sep 2014 22:42:59 +0000 https://communities.curl.com/showthread.php?tid=1107
[Image: Curl_HTML_CintaNotes_kanji_sqr_750.png]

[Image: Curl_HTML_CintaNotes_kanji_sqr_750.png]

My recent URL's include

Conning Halpern applet
Halpern Conning applet
Unicode sorted Conning-Halpern kanji]]>
[Image: Curl_HTML_CintaNotes_kanji_sqr_750.png]

My recent URL's include

Conning Halpern applet
Halpern Conning applet
Unicode sorted Conning-Halpern kanji]]> <![CDATA[Curl for the new Kodansh kanji books]]> https://communities.curl.com/showthread.php?tid=1105 Sun, 07 Sep 2014 11:51:46 +0000 https://communities.curl.com/showthread.php?tid=1105
The latter now includes Unicode values for 3002 kanji as well as index numbers for those found among the 2300 kanji of the Conning book. The course book has a few kanji not found in the learner's dictionary, but they are a very good match for someone who wants to read, say, the blog posts here in the Curl Global Community without relying on Google translate !

I have added a KanjiRecog blog and a Kanji Without Romaji page at Facebook in addition to a new CentOS virtual server with NodeJS + nunjucks templating creating the Curl kanji pages.

My concerns remain SVG for kanji, ruby characters for Furigana and seeing the RTE on Android !

次の時間まで

]]>
The latter now includes Unicode values for 3002 kanji as well as index numbers for those found among the 2300 kanji of the Conning book. The course book has a few kanji not found in the learner's dictionary, but they are a very good match for someone who wants to read, say, the blog posts here in the Curl Global Community without relying on Google translate !

I have added a KanjiRecog blog and a Kanji Without Romaji page at Facebook in addition to a new CentOS virtual server with NodeJS + nunjucks templating creating the Curl kanji pages.

My concerns remain SVG for kanji, ruby characters for Furigana and seeing the RTE on Android !

次の時間まで

]]> <![CDATA[MS OS head on ARM in the future of Windows]]> https://communities.curl.com/showthread.php?tid=1068 Fri, 11 Apr 2014 12:58:35 +0000 https://communities.curl.com/showthread.php?tid=1068
http://www.zdnet.com/microsoft-os-chief-...000028286/

in case you missed this.

The interview is with Microsoft's Executive Vice President of operating systems, Terry Myerson.]]>
http://www.zdnet.com/microsoft-os-chief-...000028286/

in case you missed this.

The interview is with Microsoft's Executive Vice President of operating systems, Terry Myerson.]]> <![CDATA[Curl on Facebook - another page]]> https://communities.curl.com/showthread.php?tid=1035 Fri, 31 Jan 2014 17:29:59 +0000 https://communities.curl.com/showthread.php?tid=1035 https://www.facebook.com/kanjirecog

the page is still hoping for 18 more LIKE visitors ...]]> https://www.facebook.com/kanjirecog

the page is still hoping for 18 more LIKE visitors ...]]> <![CDATA[large TocDocument in 64-bit browser]]> https://communities.curl.com/showthread.php?tid=1030 Sat, 18 Jan 2014 16:14:34 +0000 https://communities.curl.com/showthread.php?tid=1030
However, these same Curl files throw up a Curl debugger and fail with a stack overflow in any 32-bit browser on Windows XP.

Does anyone have a URL link to technical reference which would indicate the basis for what I am finding with these applets ? The RTE is the Caede 2.1 which I call "0.0.4" and the IDE shows as Curl Pro/IDE 8.0.1004 ( the RTE Curl manager gives

Curl® Runtime Environment
Build: 8.0.4

)

Thanks.

I am using the large TocDocument's for my own research, so this is not a "question" to which I need an answer. Wink

I am not using the Curl IDE to assemble these long Curl documents ... just a good old programming editor (not eclipse, or called from eclipse) Sleepy

as I am busy making various encoding and formatting changes of disparate literary texts in Japanese. These materials may end up being processed by Rebol "Red" or Rebol3 or ObjectIcon to get away from REGEXP without going as far as Logtalk + Prolog before mutating into epub3.

There is, of course, no "problem" here that loading PCurl's and/or dynamic options in Curl cannot solve.

Robert
Canada
UTC -4h
]]> <![CDATA[Curl for CoffeeScript folks]]> https://communities.curl.com/showthread.php?tid=925 Fri, 28 Jun 2013 17:52:12 +0000 https://communities.curl.com/showthread.php?tid=925
Why are we not getting some Curl traction with CoffeeScript developers ? Or are we ? ]]>
Why are we not getting some Curl traction with CoffeeScript developers ? Or are we ? ]]> <![CDATA[Curl JavaScript bytecodes]]> https://communities.curl.com/showthread.php?tid=918 Wed, 19 Jun 2013 13:33:16 +0000 https://communities.curl.com/showthread.php?tid=918 node.js + express + nunjucks (the latter for templating.)

The Curl applet is a variable in JavaScript whose content is constructed from three parts : a nunjucks template for a curl applet, an scurl file containing the poems and an scurl file containing the text formats to use for this page.

This should give an idea of how easily a Curl xcurl file on the server could implement such an HTTP server.

Once the page has been presented, edits to the scurl files have no impact on the page presented by the server until the server is restarted.

My default extension for the nunjucks template is .ncurl

[Image: node-curl-test31.jpg]

And it is fun, too !

The test URL for the above server-generated Curl document is http://209.208.27.168/shiki

]]> node.js + express + nunjucks (the latter for templating.)

The Curl applet is a variable in JavaScript whose content is constructed from three parts : a nunjucks template for a curl applet, an scurl file containing the poems and an scurl file containing the text formats to use for this page.

This should give an idea of how easily a Curl xcurl file on the server could implement such an HTTP server.

Once the page has been presented, edits to the scurl files have no impact on the page presented by the server until the server is restarted.

My default extension for the nunjucks template is .ncurl

[Image: node-curl-test31.jpg]

And it is fun, too !

The test URL for the above server-generated Curl document is http://209.208.27.168/shiki

]]> <![CDATA[jinja2 templates for Curl]]> https://communities.curl.com/showthread.php?tid=913 Thu, 13 Jun 2013 23:33:37 +0000 https://communities.curl.com/showthread.php?tid=913 ----------------
Curl test page

We are displaying Unicode utf-8 as Unicode brace-tagged content.

This text is emitted by a Unicode jinja2 template using Japanese quotation characters as the markup delimiters within the utf-8 text source.

No Curl braces or parentheses occur in the markup delimiters.
---------------

"brace-tagged" in this case means that the delimiters are Unicode brace characters OTHER THAN our '{' and '}' Curl brace characters - and no parentheses are used in the tag delimiters with the template.

Most web server templating is aimed at HTML or XHTML, but Jinja2 templates for Python web servers are quite agnostic. All templates are themselves utf-8.

In my experiment I have used \u300c, 300D, 3014, 3015 and 3008 with \u3009 - although other possibilities are readily available using other quotation devices.

Perhaps what is not immediately obvious is that the template ITSELF can be tested as to whether - barring substitutions - one has a valid Curl applet !

In such a test run, that Curl applet appears in the browser with items such as

「tag 」

which would be populated by the template.

My next task is the generate THIS template from lighter markup such that a test run of a page of poetry would appear with tags such as

「verse01 」

「verse02 」

「verse03 」

within their respective Curl visual, graphic or markup macros such as

{poem
{verse
「verse01 」
}
{verse
「verse02 」
}
{verse
「verse03 」
}
}

This is just one step-closer to server-side Curl for my poetry markup projects !

]]> ----------------
Curl test page

We are displaying Unicode utf-8 as Unicode brace-tagged content.

This text is emitted by a Unicode jinja2 template using Japanese quotation characters as the markup delimiters within the utf-8 text source.

No Curl braces or parentheses occur in the markup delimiters.
---------------

"brace-tagged" in this case means that the delimiters are Unicode brace characters OTHER THAN our '{' and '}' Curl brace characters - and no parentheses are used in the tag delimiters with the template.

Most web server templating is aimed at HTML or XHTML, but Jinja2 templates for Python web servers are quite agnostic. All templates are themselves utf-8.

In my experiment I have used \u300c, 300D, 3014, 3015 and 3008 with \u3009 - although other possibilities are readily available using other quotation devices.

Perhaps what is not immediately obvious is that the template ITSELF can be tested as to whether - barring substitutions - one has a valid Curl applet !

In such a test run, that Curl applet appears in the browser with items such as

「tag 」

which would be populated by the template.

My next task is the generate THIS template from lighter markup such that a test run of a page of poetry would appear with tags such as

「verse01 」

「verse02 」

「verse03 」

within their respective Curl visual, graphic or markup macros such as

{poem
{verse
「verse01 」
}
{verse
「verse02 」
}
{verse
「verse03 」
}
}

This is just one step-closer to server-side Curl for my poetry markup projects !

]]> <![CDATA[CurlDoc]]> https://communities.curl.com/showthread.php?tid=908 Mon, 10 Jun 2013 17:46:20 +0000 https://communities.curl.com/showthread.php?tid=908 CurlDoc, like rdoc for Ruby, would produce Curl docs using a PEG parser ( Markdown and MultiMarkdown )

Does Curl not already have all of the pieces for a PEG parser for converting Markdown to Curl ? And then the next task ... to emit Curl code marked-down for Curl docs.

]]> CurlDoc, like rdoc for Ruby, would produce Curl docs using a PEG parser ( Markdown and MultiMarkdown )

Does Curl not already have all of the pieces for a PEG parser for converting Markdown to Curl ? And then the next task ... to emit Curl code marked-down for Curl docs.

]]> <![CDATA[Firefox 17 config issue]]> https://communities.curl.com/showthread.php?tid=904 Mon, 27 May 2013 23:01:11 +0000 https://communities.curl.com/showthread.php?tid=904 about:config fix especially for Centos 6.4 for Firefox 17.

I.e., as Curl WBardwell advised,

enter about:config as the browser address

Right-click and select NEW then BOOLEAN

enter dom.ipc.plugins.enabled.libcurl-surge.so

as the key and then accept the default FALSE as the value, i.e.,

dom.ipc.plugins.enabled.libcurl-surge.so = false

We should also note if there are issues in running CentOS 64-bit ( I have not succeeded except with CentOS 6.4 i386 ) even in VirtualBox.

The install prep notes for Unbuntu 12.04LTS should involve more than libstdc++ warning to avoid error on libcrypto.so NOT FOUND.

Are there other issues that I have forgotten ? ]]> about:config fix especially for Centos 6.4 for Firefox 17.

I.e., as Curl WBardwell advised,

enter about:config as the browser address

Right-click and select NEW then BOOLEAN

enter dom.ipc.plugins.enabled.libcurl-surge.so

as the key and then accept the default FALSE as the value, i.e.,

dom.ipc.plugins.enabled.libcurl-surge.so = false

We should also note if there are issues in running CentOS 64-bit ( I have not succeeded except with CentOS 6.4 i386 ) even in VirtualBox.

The install prep notes for Unbuntu 12.04LTS should involve more than libstdc++ warning to avoid error on libcrypto.so NOT FOUND.

Are there other issues that I have forgotten ? ]]> <![CDATA[server-side Curl issue : TocDocument]]> https://communities.curl.com/showthread.php?tid=901 Tue, 21 May 2013 10:32:47 +0000 https://communities.curl.com/showthread.php?tid=901
Ah well ... one aspect of TocDocument class not being in open source and {document-style} proc not being in open source is that the class TocTreeNode is undocumented.

In the absence of server-side Curl frameworks this was no doubt understandable. But when I am working in a framework such as server-side Smalltalk ( in this case, Seaside 3 in Pharo Smalltalk ) it is a vexation.

Today, with the emphasis on CAEDE, what can the rationale be for keeping this legacy code out of an open-source repository ?

At some point, Curl will become a topic of interest in the history of MIT CS and it will be sad if no plan is in place for a legacy code escrow if only in PDF files.

One need only look at the plight of cinema history, where far too often there was no plan for archiving negatives and preserving intact prints from the early days of cinema. Even films from the 60's now often must be cobbled together from surviving prints.

History is documents. And in the case of programming, that is code. Is there a plan for Curl? Was it even thought of at the time of the MIT spin-off ? When looking to the future, it is understandable to forget that someone in the future will look to the past ....

Example : expression-based programming language ICON will likely be preserved at Dept. of CS, U AZ , Tucson, but the papers of the late Ralph Griswold are at UMN, MPLS ....

Example : Bach, J.S. Very little thought was given to preserving the work of the "old" Bach and in some cases Bach comes down to us through sheer good fortune ... but scholars fear much was lost.

Which CS dept in Japan has a close relationship with CS at MIT ?

]]>
Ah well ... one aspect of TocDocument class not being in open source and {document-style} proc not being in open source is that the class TocTreeNode is undocumented.

In the absence of server-side Curl frameworks this was no doubt understandable. But when I am working in a framework such as server-side Smalltalk ( in this case, Seaside 3 in Pharo Smalltalk ) it is a vexation.

Today, with the emphasis on CAEDE, what can the rationale be for keeping this legacy code out of an open-source repository ?

At some point, Curl will become a topic of interest in the history of MIT CS and it will be sad if no plan is in place for a legacy code escrow if only in PDF files.

One need only look at the plight of cinema history, where far too often there was no plan for archiving negatives and preserving intact prints from the early days of cinema. Even films from the 60's now often must be cobbled together from surviving prints.

History is documents. And in the case of programming, that is code. Is there a plan for Curl? Was it even thought of at the time of the MIT spin-off ? When looking to the future, it is understandable to forget that someone in the future will look to the past ....

Example : expression-based programming language ICON will likely be preserved at Dept. of CS, U AZ , Tucson, but the papers of the late Ralph Griswold are at UMN, MPLS ....

Example : Bach, J.S. Very little thought was given to preserving the work of the "old" Bach and in some cases Bach comes down to us through sheer good fortune ... but scholars fear much was lost.

Which CS dept in Japan has a close relationship with CS at MIT ?

]]> <![CDATA[Curl and QNX]]> https://communities.curl.com/showthread.php?tid=891 Wed, 24 Apr 2013 14:38:30 +0000 https://communities.curl.com/showthread.php?tid=891
It seems odd that Blackberry Z10 still runs Flash but innovators wanting to compete with Samsung were not taken with Curl !

Think of the many productivity app's that could have come as default or in the Blackberry store !

Productivity hounds would not be the only ones to miss out : Curl on QNX should have been a natural fit to the dual persona's of Personal and Work on the Z10.

Or am I mistaken ?

達

]]>
It seems odd that Blackberry Z10 still runs Flash but innovators wanting to compete with Samsung were not taken with Curl !

Think of the many productivity app's that could have come as default or in the Blackberry store !

Productivity hounds would not be the only ones to miss out : Curl on QNX should have been a natural fit to the dual persona's of Personal and Work on the Z10.

Or am I mistaken ?

達

]]> <![CDATA[HTML5 alternative?]]> https://communities.curl.com/showthread.php?tid=886 Wed, 10 Apr 2013 12:41:19 +0000 https://communities.curl.com/showthread.php?tid=886 Revolution
I will report back if I can get an Android app built from this "Community Edition" of their commercial IDE.
]]> Revolution
I will report back if I can get an Android app built from this "Community Edition" of their commercial IDE.
]]> <![CDATA[WikizPad Curl personal wiki]]> https://communities.curl.com/showthread.php?tid=871 Sun, 10 Mar 2013 16:09:42 +0000 https://communities.curl.com/showthread.php?tid=871 WikizPad post.

This is long over-due ... especially given the painful limitations of so many personal wiki implementations.

And Curl can address some of the security concerns that are inevitable in this kind of GTD tool.

Anyone want to help vet a wiki markup MustHave list ?

As in WikidPad, we will have SQLite and non-sql.

The Curl templating features are fun ... now why isn't this a 2013 GSoC project ? 2014 ? A project linked to code.org ?

Our app will move to Android with more ease than WikidPad --> WikiAndPad has, right CAEDE gurus ?

]]> WikizPad post.

This is long over-due ... especially given the painful limitations of so many personal wiki implementations.

And Curl can address some of the security concerns that are inevitable in this kind of GTD tool.

Anyone want to help vet a wiki markup MustHave list ?

As in WikidPad, we will have SQLite and non-sql.

The Curl templating features are fun ... now why isn't this a 2013 GSoC project ? 2014 ? A project linked to code.org ?

Our app will move to Android with more ease than WikidPad --> WikiAndPad has, right CAEDE gurus ?

]]> <![CDATA[latest flashcard markup example]]> https://communities.curl.com/showthread.php?tid=870 Sat, 09 Mar 2013 17:18:12 +0000 https://communities.curl.com/showthread.php?tid=870 Curl blog with the test mark as

{EmFontSize 36pt}
{CardFontSize 18pt}
{CardBackground "blue"}
{HiLight "red"}

{FrontSide color="gold",

日本語の漢字は {KJ \u9801} がです。

}

and the result as

[Image: kanji-flashcard-03a.png]

]]> Curl blog with the test mark as

{EmFontSize 36pt}
{CardFontSize 18pt}
{CardBackground "blue"}
{HiLight "red"}

{FrontSide color="gold",

日本語の漢字は {KJ \u9801} がです。

}

and the result as

[Image: kanji-flashcard-03a.png]

]]> <![CDATA[Curl as flashcard markup]]> https://communities.curl.com/showthread.php?tid=869 Fri, 08 Mar 2013 16:54:43 +0000 https://communities.curl.com/showthread.php?tid=869
[Image: kanji-flashcard-02a.png]

The RichTextArea allows me to zoom in on the pieces that I want to edit with care (select, rollup fontsize)

[ more to follow ]]]>
[Image: kanji-flashcard-02a.png]

The RichTextArea allows me to zoom in on the pieces that I want to edit with care (select, rollup fontsize)

[ more to follow ]]]> <![CDATA[evaluate risk]]> https://communities.curl.com/showthread.php?tid=865 Wed, 06 Mar 2013 14:47:21 +0000 https://communities.curl.com/showthread.php?tid=865 Would it not have been wiser to make 'evaluate' reserved ?

Yes, we can search for "{evaluate " ... but one could have been 'compute-position' and another 'eval-path' and leave 'evaluate' as eye-catching as we all know it should be.

Was this a failure in peer-review of code submissions ?

And how to keep more use of this identifier out of open-source contributions ?

]]> Would it not have been wiser to make 'evaluate' reserved ?

Yes, we can search for "{evaluate " ... but one could have been 'compute-position' and another 'eval-path' and leave 'evaluate' as eye-catching as we all know it should be.

Was this a failure in peer-review of code submissions ?

And how to keep more use of this identifier out of open-source contributions ?

]]> <![CDATA[アンドロイドのアプリケーションの機能]]> https://communities.curl.com/showthread.php?tid=863 Mon, 04 Mar 2013 13:48:36 +0000 https://communities.curl.com/showthread.php?tid=863
topic: GTD and personal wiki WikiAndPad app for Android

WikidPad is soon to release WikiAndPad for Android

The WikidPad personal wiki currently will export to HTML.

What if the WikiAndPad Android app could send a Curl export to a Caede "App-Me-SVP" server-side app ?

I will be doing Curl export as a Python add-in for WikidPad if it continues to be my personal wiki of choice ...

We would want to get the wiki markup pre-tuned (ask the user to make some choices) before hand-off to CAEDE app on server-side.

Is there any other "DCURL-my-wiki-SVP" idea out there for regular desktop OS's ?

]]>
topic: GTD and personal wiki WikiAndPad app for Android

WikidPad is soon to release WikiAndPad for Android

The WikidPad personal wiki currently will export to HTML.

What if the WikiAndPad Android app could send a Curl export to a Caede "App-Me-SVP" server-side app ?

I will be doing Curl export as a Python add-in for WikidPad if it continues to be my personal wiki of choice ...

We would want to get the wiki markup pre-tuned (ask the user to make some choices) before hand-off to CAEDE app on server-side.

Is there any other "DCURL-my-wiki-SVP" idea out there for regular desktop OS's ?

]]> <![CDATA[Missing context menu item?]]> https://communities.curl.com/showthread.php?tid=860 Mon, 04 Mar 2013 11:59:16 +0000 https://communities.curl.com/showthread.php?tid=860
On Puppy Linux running SeaMonkey browser in a (1 + n) workspace, a TOC Curl document sent to full-screen with F11 leaves no path out as the X jwm fails to detect workspace toggle keys in this scenario so as to navigatte to another workspace. With no response to F11 toggle, the best alternative looks to me to be a context menu item that comes active when browser is fullscreen to escape fullscreen by sending the req up to the browser.

The only key combo that works, AFAIK, is ALT-F6 to minimize Seamonkey

The downside is that maximize brings back SM but not fullscreen

It is a Curl ebook reader annoyance (not as serious as the lack of ruby characters for furigana, but ...)

Eventually this may emerge as a Dcurl issue. Is it already an Android issue ?
]]>
On Puppy Linux running SeaMonkey browser in a (1 + n) workspace, a TOC Curl document sent to full-screen with F11 leaves no path out as the X jwm fails to detect workspace toggle keys in this scenario so as to navigatte to another workspace. With no response to F11 toggle, the best alternative looks to me to be a context menu item that comes active when browser is fullscreen to escape fullscreen by sending the req up to the browser.

The only key combo that works, AFAIK, is ALT-F6 to minimize Seamonkey

The downside is that maximize brings back SM but not fullscreen

It is a Curl ebook reader annoyance (not as serious as the lack of ruby characters for furigana, but ...)

Eventually this may emerge as a Dcurl issue. Is it already an Android issue ?
]]>