Oh shoots, sorry Bob, I missed some of your questions. But that may be a blessing in disguise because if I had answered all of them in one post, that would be a record-breaking novel-like e-mail, wouldn't it? <grin>! Let's see now.
I'll start with those that still pertain to my explanations in the previous post. First, I just checked the maximum length of the field prompt (e.g., Last Name, Street Address, Business Email), and it's 30 characters. Second, you can enter any number, even a single digit, in the second line of the definition file, just have a number there. Third, you can enter up to 250 characters per field, although if you let a default value be entered for a field, then it can only have up to 50 characters. You can clear the default value by pressing SPACE with dots 3-5-6 (CTRL with BACKSPACE) and you can enter up to 250 characters again. Now, for your other questions: The usage called "Email" identifies the information entered into that field as an e-mail address, thus letting you select that definition file in the Address List option under the Keymail Setup submenu. Ideally, choosing a different definition file in that option will let you enter a different name and address appearing in the From header of your messages, so you don't have to retype a different address but just select the right definition file. However, this currently doesn't work. Information entered in one definition file will overwrite that which was previously entered in the other definition file. The usage called "File" is used for entering multiple lines of information, since the database fields are just for single-line entries. Further, this will allow you to type in data in the Keyword environment (i.e., all commands in the word processor will apply here), but that doesn't mean the file field (such as Notes) is saved elsewhere. It's still in the .cdb file associated to that definition file. Ordering of most parts of the field line, except for the identification number which should be at the start, does not have to be strictly followed. But the ideal order is: identification number, field type, usage, modifier, Braille grade, prompt, default value, help message. BTW, I just checked again and the help message you write in the definition file is read first before the default help message when you access context-sensitive help while on that field. Also, separating each part with commas does not cause any error messages being prompted, so don't wonder why some commas are missing. Those were just overlooked. Putting the commas where they belong, however, seems to improve the sorting of database records in the Lookup option. The structure of the definition file for the Directory of Services isn't that complicated. When I was investigating the syntax of the different definition files, the Email Folders ddf found in the Keymail folder of the Flash Disk bewildered me more, but now I've modified that, too... notice my reply separator heading is "quoted message" rather than "original message"? That's just one example. As I have mentioned in my previous post on this topic, there are two concatenation lines because the first is referred to by Keylist while the other by Keymail. Without the second, you will get an error tone or "Cannot find record in the Address List" message when you type a person's name in the To field upon writing an e-mail in Keymail. On the other hand, if your database does not contain info that will be accessed in Keymail, then you do not need to put a second concatenation line. However, if I recall correctly, when I was experimenting and studying the structure of Keysoft databases, the sorting of records didn't look right without the second concat line. So I just put it in, similar to the first except for the identification number, of course and that seemed to have fixed it. Again, unlike with other field lines, you must use 24594 and 24595 as the identification numbers of the two concatenation lines of a definition file to be accessed in both Keylist and Keymail. Otherwise, it's all right to use a different number. The History definition file is a system file of sorts, like the .cdb files. That's why you can view them in the File Manager, but not in Keyword or Keybook. If you wish to view its contents, there's actually a simple work-around that doesn't involve getting frustrated over unprotecting, renaming, and whatever you may have in mind of doing, LOL. 1) Enter Keymail and press W to write a message. 2) Enter any e-mail address (yours, for instance), in the To field. 3) Navigate to the Attachments field and type Y for yes when asked about attaching a file. Select the History definition file in the Keylist folder of the Flash Disk. Make sure you are viewing all file types. 4) You will be asked if you wish to attach it in another format. Just type N for no. 5) When asked if you wish to attach another file, press SPACE with dot 1 (READ with U) to move back to the first attachment. Hit ENTER, then R to open and read it. Voila! You can read it, copy it to the clipboard and paste it in a Keyword type text document, or use the Store Block command to save it in a file that's not hidden. For your reference though, here is that file's contents. database IEHistory 0 1, G0, "U R L" 2, TimeAndDate, "Last visited" 3, G0, "Domain" 4, G0, "Title" 5, array, "Cursor Position" 6, uword, "Times visited" 101, G0, "Date Domain Time Sort" 102, G0, "Domain DateTime Sort" announcement order 2, 3, 4 sort order 101, 102 Notice that the lines numbered 101 and 102 are like the concatenation lines in the Address List. Remember that you can view the browser history, sorted either by date or by domain. The default of sorting by date follows line 101. The records are sorted by date, regardless of the time. Then it is sorted by domain (e.g., all visits to Bookshare on December 14, 2003, then all visits to Google on the same day; but actually, the domain entry being sorted is a numeric address, so it's not an alphabetical ordering). And finally, the visits to a domain on a certain day will be sorted according to the time of visit. The sorting of records by domain follows line 102: domain first, then the time and date (that's why there's no space, because the time-and-date info is considered together in this case). The word "array" refers Keysoft to information derived from outside. Cursor position, for instance, is determined from the web. You'll also see the word "array" in the Email Folders database when referring to headers (since info is again gathered from the server, not something that the user enters or can just find out from some readable file). Uword is a field type where the information entered is a number between 0 and 65535. I hope that covers everything you asked. I was hurriedly typing my other post so I didn't realize I forgot to answer the questions I've answered here. Now, in fairness to PDI, even though I learned most of the stuff about Keysoft database syntax and structure through self-study and experimentation, I also learned some things from asking Dean Jackson who, in my humble opinion, does a great job as tech support specialist, as well as Maryanne Hughes Cathro, the knowledge manager. Thank you to both. Finally, I'm not sure if the respect for my writing that you mention in your message is due to the quality or quantity/length. Oh well, I suppose I'm more notorious for the latter than the former. Thanks anyway. Whenever I can find the time, I'll help. Regards, Roselle >----- QUOTED MESSAGE ----- >Sent by: "bob" <[EMAIL PROTECTED] >> Hi Listers and (especially PDI): >> line 1: database Address list I presumes this identifies this as a >database and names it. >> line 2: 24 I have no idea what this number means, perhaps the number of >fields to be defined? >> Line 3: 14865, (a unique number used later for concatenating and sorting >fields) name, (a field type with unique parameters such as length [I >suspect] autocap, (I suspect this tells the system to make sure the first >letter is capitalized whether the user capitalizes it.) "Last name" a >prompt to be used for this field. >> . >> . >> . The next lines are similar to the one just described. >> . >> . >> 14857, phone, (this is obviously a phone number type field. [What are its >restrictions?] autonum (I presume this restricts input to numeric data.), >"Home phone number" (Just another title for the heading list.) >> . >> .(More of the same) >> 16515, email, (presumably this edits the field for things like an at sign >and maybe a period) G0, (I presume this forces the user to use computer >braille for data entry) "Home email address" >> . >> . >> 16448, address, (I can presume that the address type is like the name type >except probably it allows for a longer field) "Street address" >> 16449, autocap, city, "City" (the unique thing here is that the autocap >comes before the field type city. Why? Does position matter? >> 16450, allcaps, state, "State or county" (I can presume that `allcaps` >forces the field to be in upper-case. Once again, why does the allcaps come >before the "state" field type, and what are the restrictions on this type?) >> 16451, spell, "Zip or post code" (The difference here is that the data is >spelled when read, I presume.) >> . >> . >> . >> 16393, file, "Notes" (This apparently puts input data into a file. Where >is it? >> 24594, concat, "14865, 14854" (I presume that this field is a >concatenation of the first and last name.) >> 24595, concat, "14854, 14865" (This appears to be a repeat of the above >line. Why are there two of them?) >> announcement order 14854, 14865 (I am guessing that this is the order >fields are announced in when perusing the list. Right?) >> sort order 24594, 24595 (This appears to be telling the system to sort the >database on the two concatenated fields above. Not sure why since they >apply to the same fields.) >> ----- end of guesses about the address list definition file ---- >> Things seem to get a little fuzzier when you examine the definition for >the >> list of services database. >> Finally, my paranoia is really fed when I try to look at the history >definition file (I presume this is for the iehistory list.) This file is >protected. I can unprotect it. But, try as I may, I can't get the >wordprocessor to open it. I suspect it's a hidden file and the >wordprocessor doesn't seem to look at hidden files. Why is this the case? >> Why, you might ask yourself should I take so much time macro analyzing the >system? Indeed I do have a life outside of work which has nothing to do >with the braille note. I also know that some of us users would like to do >things with our notetakers that Pulsedata never dreamed of. After all, not >all the creative people work for PDI. I am reminded of fairly recent >history when Hentor-Joyce released their scripting language for JAWS. >Immediately after that, there were a wealth of scripts coming from >users--things that have-just never anticipated. PDI come, if they dared, >have a plethora of free employees improving their system--only if they would >trust that blind people can be creative too. >> Okay, this is the end of my raving. >> I would be interested in what other users think? Even if you never plan >on creating a different database from those provided, wouldn't you like it >if others did and then posted them on this list? >> Since I believe this message is the longest I've ever written to the list, >I think it might rival some of Roselle's best efforts, and I've gained a lot >more respect for her tenacity in writing some of her missives. >> Please tell me what you think? >> And, PDI, could you comment on my analysis of a database and correct any >gross assumptions I have had to make? >> Thanks, >> Bob
