Clearwin+ documentation

Clearwin+ seems to be documented almost wholly through FTN95.CHM. This is extremely useful for looking up the details of what you already know, but it isn't the best medium for browsing for ideas. I've noted from recent posts that others find FTN95.chm difficult to use for self-instruction.

I have a copy of the thin booklet dated 1995 named 'Clearwin+ - Fortran Edition', but I also have a copy of the identically-named 450-page PDF file, dated 2000. I printed it out once, and it took a whole ream of paper -and just fits in a lever arch file and is 6220 k long). When read in conjunction with the Clearwin.enh (enhancements) file (156k long, and noted as being for v4.70), it is a source for ideas.

Some time ago, Paul put the FTN77 documentation on-line because there were several requests for documentation on library functions that still exist in FTN95 but are not documented. May I suggest that the Clearwin PDF goes there too? (In the case that Silverfrost don't have it to hand, do you want it?)

Eddie

I aggree with Eddie: Once one has mastered ClearWin+ to some extend the FTN95.CHM is very good. However, mastering ClearWin+ with FTN95.CHM is a very difficult procedure! It just takes too much time putting together the bits and pieces.

We have looked at the PDF that you mention and I cannot see any differences be the PDF and the help file. However some people find a hard copy easier to read.

Will it be possible to make the PDF on-line available?

Hi Paul,

Exactly the point. I think it would be helpful to have the PDF online. I agree that everything is in FTN95.CHM. Indeed, it was probably compiled from the original manual.

The issue is not the completeness of FTN95.CHM, but how it isn't a format that lends itself to reading 'from cover to cover' in a general quest for self-improvement!

Eddie

Unfortunately we don't have any information about the book (except a hard copy). When the transfer from Salford Software to Silverfrost occurred we took over the source code but, it appears, not all of documentation.

I am not sure whether I should continue here or perhaps in the 'Hello ClearWin+' thread. Anyway, working through the tutorial that Eddie gave me, I really started to understand many stuff previously discussed in the forum - and just by reading a few pages!

Eddie, if you don't mind we can start a tutorial in the forum based on the structure you have used - that was excellent. I volunteer to take the factoriser example and apply exactly the steps that you used. The subjects of the diffrent threads could be like this: 1.) ClearWin+ Totorial: Introduction 2.) ClearWin+ Totorial: Get started 3.) ClearWin+ Totorial: ... Doing it this way we have feedback from other users and I can take care of typesetting the the final product.

Eddie

I missed the last line of your first post - if you could send the pdf to ftn95@silverfrost.com that would be great

Thanks to Wilfried we have added the Clearwin+ Fortran Edition to the documentation page (and an old copy of the FTN95 manual in PDF format):

http://www.silverfrost.com/23/ftn95/support/documentation.aspx

Robert,

Wilfried beat me to it! That is the manual - my copy is identical. I suspected you didn't have it easily to hand at Silverfrost.

We used to be able to buy a copy of Clearwin as an add-on the FTN77 PE, and that is where mine came from. This manual is much better for browsing than FTN95.CHM, so thank you for putting it up in the Documentation section, and I'm sure that a lot of people will appreciate having it online.

I checked the old Clearwin.enh file I had. Item 184 (newest is last in the file) is about the introduction of %bk, and %bk is covered in the PDF, so I guess Wilfried and I had a late version of it.

Eddie

Thanks for the online document on ClearWin+. I found a hardcopy - dated 1995 - in one of our shelves.

Paul and Eddie

I printed out the factoriser example and (again) worked through it. However, Eddie, your tutorial is unbeatable. It will be very difficult to improve on this. I tried to put toghether a tutorial and have Paul a look at it - no way. Moreover, I really think that we should get this on the web! It addresses several (real) application components, e.g. a start-up screen; aspects dealing with the menu-bar; exit properly and even how to handle a double-click on filetype of the program. We already have something which will inspire someone to experiment with Clearwin+

Jacques

I have not seen Eddie's tutorial. If someone could point me to it I will take a look at it as soon as I can.

Paul,

Where to?

It's an unfinished opus, a 'labour of love'!

Eddie

Even in its unfinished form the tutorial is commendable! I succeeded in convincing myself that ClearWin+ is the way to go. I tried several other stuff in the past - and wasted a lot of time!!

I think that the primary reason for my time wasting is the fact that the other options (Matlab, VB. etc.) have the advantage that when you Google you get some code example within a few seconds - try this with ClearWin+ you get nothing. Even an attempt to start some online examples failed. I am sure that several ClearWin+ users are willing to share some of their code if a platform is available.

Bits of it should start appearing in Google soon: http://www.silverfrost.com/ftn95-help/clearwinp/clearwin.aspx

'It' being the Clearwin+ guide.

Seems like the text is there already, just the Figures aren't online. Can I make the suggestion that some of the illustrations are regenerated with 1 24 default.manifest as well as in raw form? This would show Clearwin+ off rather better.

Eddie

[URL=http://img8.imageshack.us/i/capture21012011173521.jpg/][/URL]

Uploaded with [URL=http://imageshack.us]ImageShack.us[/URL]

Some problems with ClearWin+, mostly with documentation.

1. There is a logical function BROWSE_FOR_FOLDER@. Yes, it does work! But only for existing folders, you couldn't add a new folder (as in standard Windows dialog). Why?

2. Help file says that you can use SET_BROWSE_FOR_FOLDER_INITIAL@ subroutine to provide an initial PATH for function BROWSE_FOR_FOLDER@ (on the page describing this function), but no details are given. My attempts were as this:

   SET_BROWSE_FOR_FOLDER_INITIAL@(init)

The answer from FTN95:

*** Statement not recognised

More funny result was with this arrangement:

res = SET_BROWSE_FOR_FOLDER_INITIAL@(init)

The answer was:

*** SET_BROWSE_FOR_FOLDER_INITIAL@ is a SUBROUTINE so cannot be used as a

Well, he is cheating! He does know what SET_BROWSE_FOR_FOLDER_INITIAL@ is, but deliberately lies! :lol: It seams that this subroutine is dropped...

That's not all my problems, but I am struggling with the rest of them and hope to find some idea to solve.

Vladimir,

Did you forget the 'CALL'? (It happened to me earlier today!)

The routine works for me:

       CHARACTER*(129) PATH(4), PROMPT
       LOGICAL DIDIT
       PATH(1)='C:\\WINDOWS\'
       CALL SET_BROWSE_FOR_FOLDER_INITIAL@ (PATH(1))
       PROMPT = 'Select default folder for datafiles'
       DIDIT = BROWSE_FOR_FOLDER@ (PROMPT,PATHWAY)
       IF (DIDIT) PATH(1) = PATHWAY

... so I'm confident that it works.

You might also find this useful. Get the program location with:

       CHARACTER*(129) PNAME
       CALL GET_PROGRAM_NAME@ (PNAME)
       CALL UPCASE@ (PNAME)
       N = LEN_TRIM (PNAME)

then you take the EXE name off the end of the string, and you have the path to your application's folder! (You need this sometimes if the program is started from a file association when you click on a data file. In this case, the working directory is not where the program is stored).

Edit 1: I've never added a folder, but there is a routine MKDIR@ (see in the FTN95 library, File manipulation routines). You may need ATTACH@ and CURDIR@ as well.

Eddie

LitusSaxonicum

Shame on me! My brains were out yesterday night! Yes, it works!