May 2008 - Oberon - mailman3.inf.ethz.ch

Re: Text2HTML.Mod (Edgar Schwarz)
by Paul Reed 31 May '08

31 May '08

Sooo... You can use Text2HTML on itself and check that into sourceforge as well, right? So we don't need an Oberon system to look at your code? :) Cheers, Paul > Message: 1 > Date: Wed, 28 May 2008 15:40:20 +0200 > From: "Edgar Schwarz" <edgar.schwarz(a)ericsson.com> > Subject: [Oberon] Text2HTML.Mod > To: "ETH Oberon and related systems" <oberon(a)lists.inf.ethz.ch> > Message-ID: > <4055256AED9D224D9442B19BF1C4C4900223A449(a)esealmw118.eemea.ericsson.se> > > > Content-Type: text/plain; charset="us-ascii" > > > Hi, > Have a look at > http://obedav.svn.sourceforge.net/viewvc/obedav/contribs/Text2HTML/ > > > Edgar

1 0

Re. Text2HTML.Mod
by Chris Glur 29 May '08

29 May '08

Edgar wrote:- > Have a look at > http://obedav.svn.sourceforge.net/viewvc/obedav/contribs/Text2HTML/ > Thanks, now it tests OK. As you will remember, 3 files are generated: 1. the actual [hot-linked] text, 2. a hot-linked contents file' 3. a hot-linked index file. After you hyperlink to a place in the text-file from an anchor *in the same text-file*, and then "Back" via the frame-menu-Button, I would expect to jump back to after the anchor/hot-link and not to the contents file. Ie. AFAIK when hot-links return they should 'return like subroutines do' ? Besides there's a direct hot-link to the contents file at the top of all 3 *.html ? Also I note:- we've got the doco with the syntax description: Tutorials.html ; which we can't use without the source-code; which sourceforge has and apparently sourceforge hasn't got the syntax description. Also I think it's good for such archives to have facilities for feedback from casual users, to show the difference between a package which has been in store for 10 years and never had a successful user, vs. one that's been downloaded and run often. Also the auto-generated FTPTool which ftp's the 3 generated files to fischer@ics and then deletes them locally, seems a bit of a cheek ! Was this Andre's idea of users sending their hand-books to him for "library centralising". Thanks, == Chris Glur.

1 0

Syntax for hyperlinked 'Book' ?
by Chris Glur 28 May '08

28 May '08

Via google-search I found that I've fallen in the same hole as Peter Easthope did 6 years ago when he wrote:- > I've studied the tutorial about Text2HTML but have > been unable to find the commands. Was Text2HTML > decommissioned? Since http://www.oberon.ethz.ch/ethoberon/tutorial/Tutorials.html tells how to make a nice hyperlinked document PLUS: "With the provision of special tags the converter offers in addition the automatic construction of a table of contents and of an index." Then after you've spent time studying it you find that Text2HTML.* would be needed, but is not available. But there ARE such nice hyperlinked documents with table-of-contents and index, called *.Book Which apparently are manipulated with BooksDev.Panel & BooksUser.Panel But the syntax to write these 'books' seems not given anywhere. The existing *.Book/s show how usefull this could be; eg. for the Regul project where the documentation is presently scattered over: Regul.Str, Regul.Tool, RegulDocumentation.Tool, RegulLicense.Text, RegulQuickstart.Tool, RegulSemantics.Tool, RegulSyntax.Tool ..at least ! Q - can somebody please tell where to find the syntax to write a text which is then compiled to give such a <hyperlinked>.Book ? So a *.Book on how to use Books* is needed; which could be created as the first exercise once the syntax is known. Thanks, == Chris Glur.

2 1

Re: Ideas needed for debugging-puzzle.
by Chris Glur 28 May '08

28 May '08

> cg> Since I transfer all my important/difficult text to S3-Oberon, ... > > I've never used S3. Doesn't it predate NO? > Is there any advantage in using S3 rather > than NO? There's a doc. which shows the Oberon family tree. But I thought native was a descendant of S3; and I want to acknowledge the existance of V4, by mentioning S3. I've got S3: NO-standAlone, DOSbased and LinuxBased; and LinuxBased-V4. ..snip... > VinciDocs.Open Manifold.Vinci > or open Manifold.Vinci in a Text viewer, mark it * > and do VinciDocs.Open * . > Wow ! Quiet amazing. It would be problematic for me, because I don't 'browse' [window shop]; I expect computers to work [or not - immediately]. So I'd miss the picture which is at the bottom of the frame and hidden from view, for Test1.Vinci . > I have more sophisticated diagrams which are > not posted for propriety. As long as you do not > disclose it, I can send one. Thanks, but I first need to master the elementary. == Chris Glur.

1 0

Re: [Oberon] Ideas needed for debugging-puzzle.
by peasthope＠shaw.ca 26 May '08

26 May '08

Chris, cg> Since I transfer all my important/difficult text to S3-Oberon, ... I've never used S3. Doesn't it predate NO? Is there any advantage in using S3 rather than NO? Thanks, ... Peter E. -- Desktops.OpenDoc http://carnot.yi.org/ Desktops.OpenDoc http://carnot.pathology.ubc.ca/ Desktops.OpenDoc http://members.shaw.ca/peasthope/

1 0

Ideas needed for debugging-puzzle.
by Chris Glur 26 May '08

26 May '08

Perhaps I shouldn't have opened this can-of-worms ? Since I transfer all my important/difficult text to S3-Oberon, I was thinking that the much Latex-format stuff eg., \exists A^{n+1} \forall x^n [x^n \in A^{n+1} \leftrightarrow \phi(x^n)] which increasingly comes on wikipedias [fetched by lynx], could easily be translated and rendered by adding the extra symbols which are not already in Math*Scn.Fnt . FontEditor.Panel would be used. I have to query the strange design decision that chose to lay-out the 256 [=16*16] key-buttons in a 20 by 13 rectangle, which leaves the last 4 places free. This despite the fact that the chars < hex21 and > hex97 are shown as 2 hex-chars on their 'buttons'. What do they teach them at ETHZ, or is it related to the fact that the French call "99" 'four twenty ten nine' ? On my linux-ETH-Oberon FontEditor.Panel doesn't display the labels on the 256 buttons, like it correctly does on my other N-O systems. Also when I copy a 'good' FontEditor.Panel to linux-ETH-Oberon, this too doesn't display the labels. Q - is this correct: a panel is an object; essentially data, showing the identifier and parameters and the object's sub-objects -- recursively. Eventually all the 'leaves' of the tree must be accessible from file. Q - I think they are kept in a library ? What is the name of the library/s ? Searching *Scn.Fnt in *.Panel shows that the texts appearing in any panel corresponds to the search-found-listed *Scn.Fnt FontEditor.Panel uses only Oberon10.Scn.Fnt, which is present in the lixux-oberon system. And all other panels look ok. Q - the faulty rendering of FontEditor.Panel is NOT related to FontEditor.Mod ? Q - Is it viable to, as a first learning exercise, use LayLa to make a 16*16 key-button-matrix, and just imitate the other ok buttons/gadgets, and just use Columbus to 'examine/transfer' the old gadget-associations ? Q - since I don't want to just replace the error without knowing the error cause, is it possible/easy to disassemble/decompile the panel into its sub-gadgets ? Thanks for any feedback, == Chris Glur.

1 0

installing the extant crazy Aos.
by PETER EASTHOPE 07 May '08

07 May '08

Folk, Finally today I had an opportunity to tackle the crazy Aos. According to AosSetup.Text and my vague memory of various annecdotes, the boot loader is BBL.Bin. Yet AosInst.Tool mentions OBL.Bin and only that is on the Aos FTP Server! Booting produces this message. Oberon loadingBad BBL.Bin According to AosSetup.Text the report should be Bluebottle loading This is confusing at best. =8-| Any ideas? Thanks, ... Peter E. -- http://carnot.yi.org/ http://members.shaw.ca/peasthope/

1 0