The documentation as currently exists has all been generated automatically starting from the FORTRAN source apart from this page.
For each routine there is a writeup page, this consists of a 'template' to be filled in by the author. SUBROUTINE / AUTHOR / DATE / LANGUAGE are mostly already filled in automatically. The major missing item to be supplied is: SHORT SUMMARY of PURPOSE: This should be a readable description. The listing and description of I/O variables/arguments also need to be completed. For these I have inserted the list of routine arguments (if any) in both INPUT and OUTPUT slots to start the you off.
I have then added further things to the template. One is called CALLING TREE from routine. This gives a list of all routines called by routine, as links to the writeup page of the routine. Plus a link to the calling tree of the routine itself. This is a pointer to a position lower down in the writeup page itself. Finally I have added a list of internal functions used by the routine also as pointers.
Finally following the template is a pointer to a file that contains all the comments that were found in the FORTRAN source. Plus all CALLs to external routines with Forward Tracker routines as links. In addition all references to Forward Tracker banks are also included as links. This file maybe useful for the author to complete the missing items. All of this should be deleted when the author has filled in the template and the link removed from the writeup page.
Use the index of routines to see authorship and so see which routines you have to supply descriptions for.
I have some further ideas of things to do in an automatic fashion, e.g. provide a list of external routines called, a list of common blocks used in the form of links. Authors would then have to provide one writeup file per common block that would describe all the variables in that common block. I also have an idea to take the CMZ tree and generate a HTML page of links. Since I want to do some further development, I have tried to think how people could start doing the missing documentation before I have completed the automatic generation of all the writeup pages with enhancements. It occurs to me that we should have the missing items also in the FTREC CMZ file and then the whole thing could be generated automatically. For example the missing pieces could go into the corresponding routine with a +SELF,IF=HTML. and I would probably need some flags for input into my program to tell me which section of the template it belongs to, but then the documentation could already start to go to SB for incorporation into the CMZ file. For example a line *HTMLP before the "Purpose" part of the text, *HTMLI before INPUT definitions and *HTMLO before OUTPUT definitions. But we should probably have further communications and discussion before anyone starts implementing anything into CMZ. Not that I think for a moment that this will happen without some additional pressure.
The name of the routine is itself a link and this gives access to the full FORTRAN source of the routine. I also have ideas for improvements in the FORTRAN part of the documentation with internal pointers to statement labels. The links to routines within a writeup are to the corresponding writeup and from within the fortran to the corresponding fortran.
Although there is an index of FTD banks, the corresponding writeup files are not yet there. I plan to write a program to create bank writeup files starting from the information that is currently in H1BANKS and H1TEXT. When this is done we will have a much better idea of what is missing. I have some other things to do so this development will go on hold for a short while. However what now exists has been done with a quick program that may well still contain bugs. e.g. Conditional CALLs are currently missed. If you see any features that are wrong, or suggestions for changes and improvements please make a list and let me have it.
Within the H1 reconstruction program H1REC the forward track reconstruction and fitting is preformed by a call to FTREC
Full alphabetical index of