ZHNCF101.NCF (Beta) Bug/Fix Report 04/03/90 Tony Parker ZHNCF101.NCF BETA TESTERS! PLEASE READ THE FOLLOWING MSG's RE:ZHNCF101! Msg 269 is 18 line(s) on 04/01/90 from HOWARD GOLDSTEIN to TONY PARKER about ZHNCF101 I just took some time to look at your ZHNCF101 file. It's a very interesting idea. I'm not aware of anything like it having been released before. However, I have noticed two problems that I think need to be addressed before this can be truly useful. 1. Your use of <1 on the -K command causes the krunched library to be built with a directory just large enough to hold all the files in it. This is fine, except that you're going to have problems the next time you update the library because there won't be room to add new members. 2. Your use of the wildcard open for the update libraries assumes that these libraries will be in chronological order in the directory. Thin about what would happen if they were not and two upadate libraries contained updates for the same member. (This is quite likely when a program is developing quickly.) In that case you could not be sure which update would end up in the master library since the -W option opens the files in the order in which they appear in the directory. I think you need to open each update file explicitly. ----------------- THIS MY REPY TO HAROLD GOLDSTEIN (04/03/90 TP) Thank you for looking at ZHNCF101.LBR; your comments will be incorporated in ZHNCF102.LBR, by changes and by inclusion in a file ZHNCF102.MSG. In answer to your comment "1)", In previous versions of ZHNCF, I did in fact set the maximum number of library members to "<128", but due to some flight of fancy on my part, I changed it to "<1"; as of ZHNCF102.NCF it goes back to "<128". I also agree with your comment "2)", the "-W" Wildcard command is a powerful but "dumb" command and there is a potential danger of overwriting files with the same name, however for the purposes of updateing Bill's Z3HELPnn.LBR files that becomes an advantage. I will however include into the ALIAS in ZHNCF102.NCF a command to invoke SAP60, to clean up the directory first. I agree with you that use of explicit library and library member names is the "safest" procedure with "-W". BTW, how would you suggest "passing" of parameters from the system to NULU? I'm trying to come up with some ZEX, ZFILER, and ZMANG scripts, so please keep the comments and suggestions coming Thank you, TP. ---------------------- THE FOLLOWING ARE THE TWO BUGLET FIXES!!!!! 1.) There is a slight buglet in ZHNCF101.NCF which Howard spotted an can be very quickly corrected; ZHNCF101.NCF is an ascii file so bring it up with a word-processor in either ascii or non-doc mode and do a MASS "find and replace" ("^QA" in Wordstar or VDE) changing "<1" to "<128", make sure you get all of them 'cause it occurs several times in ZHNCF101.NCF. 2.) The second "buglet" is more in the nature of good Z-System "table manners" the NULU "-W" (Wildcard open) command opens ambiguously "filespec'd" libraries in descending order as they occure in the disk drive directory. "-W" is a powerful but "dumb" command, and if the directory is a little messy, files might get trashed or overwritten. I would recommend a policy of keeping a good control on your filenames, and to use explicit library and library member names wherever possible. It is a good practice to "clean, sort and pack" your directory with SAP60 or some other utility prior to bringing up ZHNCF101.NCF. The best place to put such a command is in the "ZHNCF101" ALIAS so it would be automatically invoked every time, so change your ALIAS "ZHNCF101" in your ALIAS.CMD file thus: ZHNCF101 d12:;a0:sap60 d:;a0:nulu -< d12:zhncf101.ncf;$hb: If you've already processed your On-Line Help System then you need to "re-Krunch" your Help libraries A.LBR thru Z.LBR allowing for a larger amount of library entries ie. change "<1" to "<128". So insert the following ALIAS in your ALIAS.CMD file thus: REKRUNCH d12:;a0:nulu -w d12:?.lbr -k <128 -l -x;$hb: DON'T FORGET to change the above D/U: (Drive/User) references to suit your system file locations in both the ALIAS's and your ZHNCF101.NCF file. Remember NULU152 does not parlez-vous "Z-System" named directories so use ZCPR2 D/U: references only! ------------ Please post any further bug reports, comments, or suggestions in the form of messages on the following Z-Nodes I call "Home": Plainfield Z-Node # 32, SysOp Chris McEwen, 201-754-9067 Manhattan Z-Node # 15, SysOp Liv Hinckley, 212-489-7370 Drexel Hill Z-Node # 6, SysOp Robert Dean, 215-623-4040 Newton Centre Z-Node # 3, SysOp Jay Sage, 617-965-7259 Thank You, TP. 04/03/90