THE WELCH COMPANY
440 Davis Court #1602
San Francisco, CA 94111-2496
415 781 5700


S U M M A R Y


DIARY: November 11, 2002 02:47 PM Monday; Rod Welch

Improve HTML conversion for transfer to Internet in 0122 and 070405.

1...Summary/Objective
2...Improve Conversion of SDS Records to HTML
3...Document Type Declaration Missing from SDS but Not from Documents
4...Letterhead for SDS Records
5...Diary Executive Summary Line
6...Diary Record and Contact Identification for Email
7...Record Description 01 Line
8...Action Item and Comment Reply Notice
9...Indented Headings Bold
10...Headings Need Only a Single End Tag for Fonts
11...Entity Renderings for & > Convert to HTML
12...Anchors and Entities Convert to Temporary Symbols for HTML Conversion
13...Single Space Anchors Above Text By Deleting Extra Blank Line
14...Single Space Below Subject Descriptions and Beginning of Narrative
15...Anchor Symbol Changed to Char 235
16...Anchors Convert to HTML
17...Double Dots Addressability in SDS Records
18...HTTP Conversion to URL for Access on the Internet
19...Distribution Ending HTML at Bottom of File
20...Lines Added to Display Anchor at Bottom of Record at Top of Browser
21...Anchor Line at Bottom of Record Displays at Top of Browser Add Lines
22...Document Conversion to HTML Modified According to W3C Standards
23...DocType Declaration Missing from Documents
24...Letterhead for Documents Converted to HTML
25...Date and Doc ID Table Corrected Font and TD Tag Problem
26...Entities in Documents Difficult to Convert to HTML
27...Address Memo Format Enhanced Process to Integrate with To From
28...Subject Table COLS Parm Seems to Reduce Flexibility of Window Size
29...Post Script Format No Spaces Below Signature Block
30...P Command Switch to BR BR to Avoid Standards Issues
31...End Document Formatting
32...Restore Original Position in Document After Conversion


..............
Click here to comment!

CONTACTS 
0201 - Dynamic Alternatives                         562 802 1639
020101 - Mr. Garold L. Johnson
020102 - dynalt@dynalt.com

SUBJECTS
HTML Improvements Validator Shows SDS Code Can be Improved 070405 012
HTML Conversion Not Displaying Correctly in Opra Browser String of Te
Problem HTML Conversion Not Displaying Correctly in Opra Browser, Gar
Validator Program Shows HTML Code Defective, Gary Johnson
Johnson, Gary Provides Examples of Specific Problems
Template Files From Gary to Enable Fixing Code Errors for HTML
Font Nesting Errors HTML in SDS Records Documents Need Correction So
Nesting Font Bold Anchor Command Errors HTML in SDS Records Documents
Templates to Help Modify SDS Avoid Nesting Errors Commands Out of Ord
Templates Used to Modify SDS Code in Order to Avoid Nesting Errors Co
HTML Code Improvements Implemented Based on Template Gary Submitted S
Planning Concept Objectives

2614 -
2614 -    ..
2615 - Summary/Objective
2616 -
261601 - Follow up ref SDS 42 0000, ref SDS 39 0000.
261602 -
261603 - Modified HTML processors so that SDS records are now closer to format
261604 - samples Gary submitted that display on the Opera browser.  Main change
261605 - for converting documents to HTML was to eliminate the P command and
261606 - use br br. ref SDS 0 ZE4F  Hope this will eliminate some format
261607 - issues.  Developed code to convert HTML programming symbols for & > to
261608 - what are called "entities." ref SDS 0 5S4J  Revised code to convert
261609 - anchors into HTML using a new char 235, so that the < symbol can
261610 - be treated as an entity. ref SDS 0 B1YQ  Reduced redundant end tags.
261611 - Changed routines to reduce "nesting" issues, ref SDS 0 E64J, e.g., the
261612 - order of HTML that ends an SDS record. ref SDS 0 0O6I  Added new
261613 - feature to add lines at the end of a record so that the last anchor
261614 - displays at the top of a browser. ref SDS 0 F874  Improved system of
261615 - formatting memos that use From and To fields rather than conventional
261616 - letter address. ref SDS 0 I54L  Corrected table format for date and
261617 - doc ID fields, ref SDS 0 4M3L; encountered difficulty using table
261618 - format for Subject field, but eventually solved it. ref SDS 0 654F
261619 - Improved restoring a document file in the original relative position
261620 - after anchors are added during HTML processing. ref SDS 0 PESS
261621 -
261622 -     [On 021114 Gary reports SDS records and documents now display in
261623 -     Opera, indicating closer compliance with HTML standards.
261624 -     ref SDS 46 0001
261625 -
261626 -
261627 -
261629 -  ..
2617 -
2618 -
2619 - Planning
2620 -
262001 - Improve Conversion of SDS Records to HTML
262002 -
262003 - Follow up ref SDS 45 TW6F, ref SDS 42 HE4U.
262004 -
262005 - Background
262006 -
262007 -    On 021102 shows background leading up this work. ref SDS 42 I74J
262009 -     ..
262010 -    On 021106 received template files from Gary Johnson for guidance on
262011 -    improving HTML code in macros 0122 and 070405. ref SDS 45 XY4E
262013 -  ..
262014 - Today, I am going to try changing the code to produce work product
262015 - that uses formatting in the template files Gary submitted on 021106.
262016 -
262017 -     a.  Original SDS record, ref DRP 5 EX8G, for 021028,
262018 -         ref SDS 41 0001,
262019 -
262020 -         Corrected version has Gary's comments. ref DRP 5 EQRQ
262021 -
262022 -             This is implemented below, and seems to run in both IE
262023 -             and Netscape.  Hopefully it will run in Opra.
262025 -          ..
262026 -     b.  Original letter, ref DRP 5 PRQV, is dated 021028,
262027 -         ref DIP 1 0001,
262029 -          ..
262030 -         Corrected version has Gary's comments. ref DRP 5 0952
262031 -
262032 -             This is implemented below. ref SDS 0 FO4O
262033 -
262034 -
262035 -
262036 -
262037 -
262038 -
2621 -

SUBJECTS
Introduction to Structure of HTML Document
Doctype Command First Line in HTML File Shows Comment on HTML Version

3004 -
300501 -  ..
300502 - Document Type Declaration Missing from SDS but Not from Documents
300503 -
300504 - Follow up ref SDS 45 XY93.
300505 -
300506 - Gary proposes adding a doc type to all files....
300507 -
300508 -     SDS records, ref DRP 5 EQRQ,
300510 -  ..
300511 - For now I am going to skip this until there is more information on
300512 - analysis in the record on 021102. ref SDS 42 JL5O
300513 -
300514 -
300515 -
300516 -
300517 -
300518 -
3006 -

SUBJECTS
Letterhead (Top of Form)

3103 -
310401 -  ..
310402 - Letterhead for SDS Records
310403 -
310404 - Follow up ref SDS 42 K97V. ref SDS 42 HE4U
310405 -
310406 - On 021102 I did some of these changes that Gary is proposing today.
310407 - ref SDS 42 K97V
310409 -  ..
310410 - The code we now produce looks like this...
310411 -
310412 -     <center><font face="Times Roman" SIZE=2><B>
310413 -     THE WELCH COMPANY<BR>
310414 -     440 Davis Court #1602<BR>
310415 -     San Francisco, CA  94111-2496<BR>
310416 -     415 781 5700<BR></B></font>
310418 -  ..
310419 - The code Gary proposes looks like this...
310420 -
310421 -     <center><font FACE="Times Roman" SIZE=2><B>
310422 -     THE WELCH COMPANY<BR>
310423 -     440 Davis Court #1602<BR>
310424 -     San Francisco, CA  94111-2496<BR>
310425 -     415 781 5700<BR></B>
310426 -     </font>
310427 -
310428 - ...in his template file. ref DRP 5 IQ6J
310430 -  ..
310431 - Line 1130, ref OF 5 624K, -label 02 in 070405 about 100 lines below
310432 -
310433 -    This is where the code for the letterhead starts, and it appears
310434 -    to be the same that Gary is proposing, so no change.
310436 -  ..
310437 - Line 1180, ref OF 5 CZ8O, -label lpltrd in 070405 about 20 lines
310438 - below
310439 -
310440 -    This is where HTML is added to end the letterhead, and appears to
310441 -    be to be the same that Gary is proposing, so no change.
310442 -
310443 -
310444 -
310445 -
310446 -
310447 -
310448 -
310449 -
3105 -

SUBJECTS
Executive Summary

3203 -
320401 -  ..
320402 - Diary Executive Summary Line
320403 -
320404 - The code we now produce looks like this...
320405 -
320406 -     <FONT SIZE=6 COLOR="red"></A><A NAME="0000"></A>
320407 -     <B>S U M M A R Y</B><BR></FONT><FONT COLOR="black">
320408 -     </FONT></center>
320410 -  ..
320411 - The code Gary proposes looks like this...
320412 -
320413 -     <font SIZE=6 COLOR="red"><B>S U M M A R Y</B><BR></font></center>
320414 -
320415 - ...in his template file. ref DRP 5 6E7M
320417 -  ..
320418 - Line 1280, ref OF 5 5813, -label lpltrd in 070405 about 120 lines
320419 - below...
320420 -
320421 -    Applied the correction Gary proposes.
320422 -
320423 -
320424 -
320425 -
320426 -
320427 -
3205 -

SUBJECTS
Diary Record ID Field
Record Description 01 Line

3304 -
330501 -  ..
330502 - Diary Record and Contact Identification for Email
330503 -
330504 -
330505 - The code we now produce looks like this...
330506 -
330507 -     <BR><FONT SIZE=5>
330508 -     <B>DIARY:</B> </FONT><FONT SIZE=4><I>October 28, 2002
330509 -     10:12 PM Monday;<A HREF="mailto:rodwelch@pacbell.net">
330510 -     Rod Welch</A></I></FONT><BR>
330511 -     <BR>
330513 -  ..
330514 - The code Gary proposes looks like this...
330515 -
330516 -     <BR><font SIZE=5>
330517 -     <B>DIARY:</B> </font><font SIZE=4><I>October 28, 2002
330518 -     10:12 PM Monday;<A HREF="mailto:rodwelch@pacbell.net">
330519 -     Rod Welch</A></I></font><BR>
330520 -     <BR>
330521 -
330522 - ...in his template file. ref DRP 5 HA8K
330523 -
330525 -  ..
330526 - Line 1290, ref OF 5 2463, -label lpltrd in 070405 about 120 lines
330527 - below...
330528 -
330529 -    Applied correction Gary proposes; actually they seem to be the same
330530 -    thing, reflecting modifications on 021102. ref SDS 42 TW6F
330531 -
330533 -  ..
330534 - Record Description 01 Line
330535 -
330536 - The code we now produce looks like this...
330537 -
330538 -      </FONT><FONT SIZE=5 COLOR="black"><B>Gary was sent SDS
330539 -      update with improvements to writing a letter.</B> </FONT>
330540 -      <FONT COLOR="black"><BR>
330541 -      <BR><FONT SIZE=4>
330543 -  ..
330544 - The code Gary proposes looks like this...
330545 -
330546 -      <font SIZE=5 COLOR="black"><B>Gary was sent SDS update
330547 -      with improvements to writing a letter.</B></font><BR>
330548 -      <BR><font SIZE=4 COLOR="black">
330549 -
330550 - ...in his template file. ref DRP 5 RP9H
330552 -  ..
330553 - Line 1350, ref OF 5 MD5J, -label g01L in 070405
330554 - below...
330555 -
330556 -      Implemented Gary's suggestion.
330557 -
330558 -
330559 -
330560 -
330561 -
330562 -
3306 -

SUBJECTS
Font for Action Items

3403 -
340401 -  ..
340402 - Action Item and Comment Reply Notice
340403 -
340404 - The code we now produce looks like this...
340405 -
340406 -      ..............</FONT><FONT SIZE=4></B>
340407 -      <A HREF="mailto:rodwelch@pacbell.net
340408 -      </FONT><FONT COLOR="red">Click here to comment!
340409 -      </A></FONT><FONT COLOR="black" SIZE=4><BR>
340410 -      <BR>
340412 -  ..
340413 - The code Gary proposes looks like this...
340414 -
340415 -      ..............</font>
340416 -      <A HREF="mailto:rodwelch@pacbell.net">
340417 -      <font SIZE=4 COLOR="red">Click here to comment!
340418 -      </font></A><font COLOR="black" SIZE=4><BR>
340419 -      <BR>
340420 -
340421 - ...in his template file. ref DRP 5 HJ5J
340423 -  ..
340424 - Line 1510, ref OF 5 QE6H, -label comment in 070304
340425 -
340426 -      Implemented Gary's suggestion.
340428 -  ..
340429 - Line 1520, ref OF 5 QA3H, -label nAIx in 070405
340430 -
340431 -      Implemented Gary's suggestion.
340432 -
340433 -
340434 -
340435 -
340436 -
340437 -
3405 -

SUBJECTS
Indented Headings 4 place nums

3503 -
350401 -  ..
350402 - Indented Headings Bold
350403 -
350404 - Line 1780, ref OF 5 5076, -label z5601 in 070405
350405 -
350406 -    The code here begins to process indented headings on 4 place line
350407 -    numbers.
350409 -  ..
350410 - Line 1820, ref OF 5 6607, -label fiTxt in 070405 about 20 lines below
350411 -
350412 -    ins_text!</font><font FACE="Arial,Helvetica" SIZE=3 COLOR="black"><B!
350413 -
350414 -       Changed this code to add a b command at the end that makes
350415 -       indented headings bold.
350417 -  ..
350418 - Line 1840, ref OF 5 M14L, -label fiTxt in 070405 about 40 lines below
350419 -
350420 -     ins_text!</B></FONT><FONT FACE="Courier" SIZE=2>!
350421 -
350422 -        Changed this to support bold on indented headings, by turning
350423 -        bold off at the end of the string.
350424 -
350425 -
350426 -
350427 -
350428 -
3505 -

SUBJECTS
Headings

3603 -
360401 -  ..
360402 - Headings Need Only a Single End Tag for Fonts
360403 -
360404 - Line 2900, ref OF 5 7933, -label bafln
360405 -
360406 -      ins_text!"></A></font></font><font FACE="Arial,Helvetica" SIZE=4 COLOR="red"!
360407 -    ins_text!"></A></font><font FACE="Arial,Helvetica" SIZE=4 COLOR="red"!
360408 -    -label uaebu
360409 -      -if @117 = 1 -if @11 = 1 ins_text!</font></font><font FACE="Arial,Helvetica" SIZE=4 COLOR="red"!
360410 -    -if @117 = 1 -if @11 = 1 ins_text!</font><font FACE="Arial,Helvetica" SIZE=4 COLOR="red"!
360411 -
360412 -       Try eliminating the double /font, so there is only one.
360413 -
360414 -
360415 -
360416 -
360417 -
3605 -

SUBJECTS
Entities < & > Convert to HTML Strings
^ Symbol Convert < to Create HTML Anchor for Linking

3804 -
380501 -  ..
380502 - Entity Renderings for < & > Convert to HTML
380503 -
380504 - Gary suggests adding code to convert the following...
380505 -
380506 -                           <  &  >
380507 -
380508 - ...common symbols that occur in SDS records and documents, into HTML
380509 - command strings to distinguish text from commands for display in a
380510 - browser.
380511 -
380512 -      [On 021116 Gary Johnson submitted ideas for converting entities.
380513 -      ref SDS 48 0001
380515 -       ..
380516 -      [On 021120 Gary pointed out a problem using ^ symbols in 070405
380517 -      to convert anchors, results in having to manage the ^ like
380518 -      another entity. ref SDS 49 5C6F
380520 -  ..
380521 - This can be accomlished in SDS records with relative ease because the
380522 - records are not created with HTML that is intended to control the
380523 - appearance of the file in a browser.  Any HTML in an SDS record, prior
380524 - to publication is there only as an object of discussion and analysis.
380525 - Therefore, before HTML is added in 070405, we can convert all HTML
380526 - control symbols that might be in the record without affecting the
380527 - intended rendering in the browser.
380528 -
380529 -      [...below, converting entities in a document is reviewed.
380530 -      ref SDS 0 494F
380532 -  ..
380533 - For SDS records this seems to work...
380535 -  ..
380536 - Line 1060, ref OF 5 TW7I, -label 02 in 070405 about 30 lines below
380537 -
380538 -      z 6 230
380539 -    z 6 170
380540 -
380541 -       Try using zone 6 170 because for Contacts and References if
380542 -       there are entities we want to avoid the change command from
380543 -       shifting links that are in column 171 off alignment, because
380544 -       then link conversion is messed up.
380546 -        ..
380547 -       The only issue is if there might be some html that pushes some
380548 -       line draw char beyond column 170.  A this time in processing,
380549 -       no HTML has been added yet, so we should be okay.
380550 -
380552 -  ..
380553 - Line 1100, ref OF 5 PY6O, -label 02in 070405 about 60 lines below.
380554 -
380555 -       Change all instances of > and & to something unique for "&" in
380556 -       the command strings for these entities.  Then change the uniqe
380557 -       stuff for & to complete the entity commands, so we wind up being
380558 -       rigged correctly with HTML strings for > and &.  Sounds
380559 -       confusing, but this is a subtle issue.
380560 -
380562 -  ..
380563 - Anchors and Entities Convert to Temporary Symbols for HTML Conversion
380564 -
380565 -
380566 - Line 1130, ref OF 5 UP4L, -label cathtl in 0704050
380567 -
380568 - Modify work on 000724 that used change command to create temporary
380569 - symbol for converting SDS anchor strings into HTML anchors.
380570 - ref SDS 29 0618
380571 -
380572 -    recl 300
380573 -
380574 -       Expand standard record length in case somthing we do is longer
380575 -       than 260 char.
380577 -     ..
380578 -    -label cathtl
380579 -    of 0
380580 -    setcnt 0 0
380581 -    l /</
380582 -    -if @0 > 0 -goto ahtml
380583 -
380584 -       Do a loop looking for < and restore the screen offset each pass,
380585 -       since the next step can offset the screen.  When all < in the
380586 -       file have been processed it is over.
380588 -  ..
380589 - Line 1150, ref OF 5 1236, -label cathtl in 070405 about 30 lines below
380590 -
380591 -    loc_cur 0 1
380592 -    strcnt 1 2
380593 -    -if @1 = 3 -goto cathtl
380594 -
380595 -        Put the cursor on col 1 and check to see if we landed in the
380596 -        Reference field, and if so, go look for the next entity,
380597 -        ref SDS 0 BX4O, because we do not need to process < in
380598 -        References because this field is removed from the published
380599 -        record on the Internet, and because if an entity is processed
380600 -        it shifts the links in column 171 off line, and then we cannot
380601 -        do HTML conversion on the links.
380602 -
380603 -    -goto lpals
380604 -        ..
380605 -        This means we are not in the Reference field, and so can
380606 -        process this as a potential anchor, so jump to code that does
380607 -        that. ref SDS 0 HD68
380608 -
380610 -  ..
380611 - Single Space Anchors Above Text By Deleting Extra Blank Line
380612 -
380613 - To obtain consistant appearance of text separation, decided to avoid
380614 - double spacing paragraphs where an anchor is above the text, rather
380615 - than inline.  Assume if text is inline, that author has set line
380616 - spacing, if anchor is above text, and the line above it is blank, then
380617 - the blank line is removed so that the anchor line separates
380618 - paragraphs. (this work was actually done on 030107, but reported here)
380620 -        ..
380621 -       [On 021228 explain experience that led to this change.
380622 -       ref SDS 50 VS7J
380624 -  ..
380625 - Line 1160, ref OF 5 1Z7K, -label anc235 in 070405
380626 -
380627 -    setlen 1 0
380628 -    subcnt 1 @8
380629 -    -if @1 > 6 -goto dndln
380630 -
380631 -       Get length of current line and subtract left margin to get
380632 -       actual length of the line, and if the line is longer than an
380633 -       anchor, it means the anchor is inline, so go switch the anchor
380634 -       symbol.
380636 -     ..
380637 -    down
380638 -    setlen 1 0
380639 -    -if @1 < 10 immed d
380640 -    -if @1 > 10 up
380641 -
380642 -       We have determined that the anchor is above the text, rather
380643 -       than inline, so check the line above.  If the line above is
380644 -       blank, delete it, and if not scroll back up to the original
380645 -       position, so that we avoid double spacing between paragraphs,
380646 -       unless the user expressly sets that condition.
380648 -           ..
380649 -          [...below, adopt same single space format to separate subject
380650 -          descriptions and narrative. ref SDS 0 K37H
380652 -     ..
380653 -    ic 235
380654 -
380655 -       Code has determined this is an anchor, so change the symbol
380656 -       from < to char 235 for identification later when anchors are
380657 -       converted to HTML.
380658 -
380660 -  ..
380661 - Line 1170, ref OF 5 118N, -label lpals in 070405
380662 -
380663 -    fr /</
380664 -    -if @0 > 0 -goto cathtl
380665 -
380666 -       Do a loop to look for all instances of < on the line, and when
380667 -       it is over, end the loop and look for < on another line.
380668 -       ref SDS 0 OW4J
380670 -     ..
380671 -    rel_cur 0 -1
380672 -
380673 -       fr puts cursor one column past the < symbol, so back up to put
380674 -       the cursor on the symbol to verify it is an anchor and not
380675 -       an entity.
380677 -     ..
380678 -    pushc
380679 -    loc_cur 0 1
380680 -    aw
380681 -    aw
380682 -
380683 -       We need to test to verify that this an anchor that should be
380684 -       processed as an anchor and not as an entity, and that requires
380685 -       determining that this < symbol is the first char on the line
380686 -       that is part of the narrative and is not part of the line number
380687 -       field. So save the cursor position on the < symbol, and move it
380688 -       to column 1, then do aw twice, because this sequence of aw
380689 -       commands will put the cursor on first text string on the line
380690 -       by jumping past the -.
380692 -     ..
380693 -    chrcnt 1 0
380694 -    popc
380695 -    -if @1 != 60 -goto ltsymb
380696 -
380697 -       Capture the char and pop the cursor back to the "<" position. If
380698 -       the first char on the line is not a "<" symbol, then the symbol
380699 -       cannot be entered as an anchor, and so go process it is an
380700 -       entity. ref SDS 0 PUUU
380702 -  ..
380703 - Line 1190, ref OF 5 9U4O, -label lpals in 070405 about 30 lines below
380704 -
380705 -    macro 1361
380706 -    rel_cur 0 -6
380707 -
380708 -       The code above has determined this is the first < symbol on the
380709 -       line, and so it may be an anchor.  Run standard anchor test
380710 -       macro and then put the cursor back on the symbol.
380712 -     ..
380713 -    -if @6 = 1 -goto anc235
380714 -
380715 -       If it is an anchor shown by 6 = 1 enter char 235 on the ASCII
380716 -       chart to distinguish HTML code from anchors for processing later
380717 -       in the operation, and go look for another < in the file.
380718 -       ref SDS 0 YX6F  Below, the code will change char 235 back to an
380719 -       HTML anchor symbol <. ref SDS 0 GF5O
380720 -
380721 -          [On 021120 Gary Johnson reports problem using ^ symbol
380722 -          temporary conversion op, so changed to char 235. ref SDS 49
380723 -          5C6F and ref SDS 49 Y76N
380725 -  ..
380726 - Line 1200, ref OF 5 WS3O, -label ltsymb in 0704050
380727 -
380728 -    ic 38
380729 -
380730 -       It is not an anchor, so must be an entity.  Change < to an & to
380731 -       identify an entity op.
380733 -     ..
380734 -    split
380735 -    right 5
380736 -
380737 -       In case we are near the right screen edge.
380739 -     ..
380740 -    ic 108
380741 -    ic 116
380742 -       ..
380743 -       Complete entity spec for <
380745 -     ..
380746 -    join
380747 -    ic 59
380748 -    -goto lpals
380749 -
380750 -
380751 -       A string that is text and not an anchor, split the text after
380752 -       the < symbol, and complete entering HTML code that identifies
380753 -       an entity for <, then join the rest of the line, and go look
380754 -       for another < to process, per above. ref SDS 0 BX4O
380755 -
380756 -
380758 -  ..
380759 - Single Space Below Subject Descriptions and Beginning of Narrative
380760 -
380761 - Line 1970, ref OF 5 587O, -label bslist in 070405
380762 -
380763 -    Just unplugged the immed d command that removed the blank line
380764 -    below the double dash that identifies the end of the lines the
380765 -    describe subjects for a record segment.  This allows Users to have
380766 -    a single blank line below the subject description field and the
380767 -    narrative, which usually begins with an anchor and a headline.  It
380768 -    complements the change to single spaces between para using the code
380769 -    above to set the format when there is an anchor between paragraphs.
380770 -    ref SDS 0 HD8O
380771 -
380772 -
380774 -  ..
380775 - Anchor Symbol Changed to Char 235
380776 -
380777 - Line 2040, ref OF 5 673L, -label 070405_2 in 070405
380778 -
380779 -      l /^0001/
380780 -    l /_0001/
380781 -
380782 -       Per above, changed the anchor symbol from ^ to char 235.
380783 -       ref SDS 0 YX7H
380784 -
380785 -
380787 -  ..
380788 - Anchors Convert to HTML
380789 -
380790 - Line 2610, ref OF 5 ZN9M, -label lpcar in 070405
380791 -
380792 -       Per above, changed temp anchor symbol from ^ to char 235.
380793 -       ref SDS 0 YX7H
380794 -
380796 -  ..
380797 - Line 2960, ref OF 5 XM6L, -label fAnC in 070405
380798 -
380799 -       Per above, changed temp anchor symbol from ^ to char 235.
380800 -       ref SDS 0 YX7H
380801 -
380802 -
380804 -  ..
380805 - Line 3310, ref OF 5 XF6I, -label over in 070405
380807 -  ..
380808 - Change code for removing Reference field so this is done before
380809 - removing links in column.  Previously, we would process the links in
380810 - column 171, then delete the field, and so there is no point to do
380811 - processing somethint that is then deleted.
380812 -
380813 -    z 1 3
380814 -    l-/03 /
380815 -    loc_cur 3 1
380816 -    immed dd
380817 -
380818 -       We are at the bottom the file, so do a reverse search to put
380819 -       the top of the Reference field on line 1; put the cursor on
380820 -       that line and start a block delete.
380822 -     ..
380823 -    z 1 1
380824 -    l / /
380825 -    down
380826 -
380827 -       z 1 1 and find the first blank line because processing has
380828 -       created a blank line below the bottom of the Reference field
380829 -       and the beginning of the first record segment.  Move down a
380830 -       line to leave the blank to separate Contacts from Subjects when
380831 -       Reference field is deleted in next step.
380833 -     ..
380834 -    immed dd
380835 -    macro 6
380836 -
380837 -       Complete block delete, and then mark this location for
380838 -       subsequent processing.
380839 -
380840 -
380841 -
380842 -
380843 -
380844 -
3809 -

SUBJECTS
2 Dots Addressability Anchors Add Links Above so People Can Copy to I

3903 -
390401 -  ..
390402 - Double Dots Addressability in SDS Records
390403 -
390404 -
390405 - Line 2620, ref OF 5 AT5K, -label enLnk in 070405
390406 -
390407 -    macro 1181
390408 -    rel_cur 0 -1
390409 -    ins_text !">..</A>!
390410 -
390411 -       Gary pointed out a few days later that there has been a space
390412 -       between the anchor string and the "> that ends the entry for a
390413 -       redundant link in SDS records.
390415 -        ..
390416 -       The correction is rel_cur 0 -1
390417 -
390418 -
390419 -
390420 -
390421 -
3905 -

SUBJECTS
HTTP Links

4003 -
400401 -  ..
400402 - HTTP Conversion to URL for Access on the Internet
400403 -
400404 - Line 3410, ref OF 5 5190, -label lp8 in 070405
400405 -
400406 -    No code changes needed for this.
400407 -
400408 -
400409 -
400410 -
400411 -
4005 -

SUBJECTS
HTML Ending at Bottom of File
Distribution Notice Based on Contacts Listed in the Record
Anchor Line Last One at Bottom of the Record Displays at Top of Brows

4305 -
430601 -  ..
430602 - Distribution Ending HTML at Bottom of File
430603 -
430604 -
430605 - Line 2820, ref OF 5 M53L, -label end in 070405 about 20 lines below
430606 -
430607 -    top
430608 -    errorcnt 0 0
430609 -    z 1 5
430610 -    l /0201 /
430611 -    bot
430612 -    immed 3a
430613 -    ins_text!</font></PRE>!
430614 -    -if @0 > 0 -goto nCn
430616 -        ..
430617 -       Gary pointed out a few days later in a letter that there is a
430618 -       nesting error with /font following the /pre command in SDS
430619 -       records on the internet.  This code puts the /font command ahead
430620 -       of /pre.
430622 -        ..
430623 -       At top of the record look for line 0201 because that shows there
430624 -       are contacts, if @0 = 0.
430626 -        ..
430627 -       Jump to the bottom of the record and provide 3 new lines to
430628 -       enter ending HTML for font and pre commands, and to enter a
430629 -       notice on distribution to Contacts, if any are in the record
430630 -       based on @0 = 0.
430631 -
430633 -  ..
430634 - Line 2840, ref OF 5 2960, -label end in 070405 about 30 lines below
430635 -
430636 -     c.
430637 -    ins_text!<font FACE="Arial,Helvetica" SIZE=3 COLOR="black"><B>!
430638 -    eol
430639 -    ins_text!Distribution. . . . See "CONTACTS"</B></FONT>!
430640 -    loc_cur 0 1
430641 -
430642 -       This means a contact is in the record beginning on line 0201,
430643 -       so give notice of distribution to contacts.
430644 -
430645 -
430647 -  ..
430648 - Lines Added to Display Anchor at Bottom of Record at Top of Browser
430649 - Anchor Line at Bottom of Record Displays at Top of Browser Add Lines
430650 -
430651 - Line 2860, ref OF 5 FT4I, -label nCn in 070405 about 20 lines below
430652 -
430653 - Added feature to automatically add enough lines at the bottom of the
430654 - record so that the last anchor in a record will display at the top of
430655 - the browser that people who open links to that anchor can find
430656 - information in the right context quickly and easily, as related on
430657 - 010916. ref SDS 31 0001  Without this feature, there is loss of
430658 - orientation and additional time required to investigate extraneous
430659 - information, with consequent potential for error, when a link is
430660 - opened, and it is not at the top of the browser, and there are one or
430661 - two other links between the top line of the browser and the actual
430662 - target line because the file does not have enough lines for the target
430663 - line to move to the top of the browser.
430664 -
430665 -     d.
430666 -    macro 20
430667 -    z 40 90
430668 -    l-/>../
430669 -    loc_cur 3 1
430670 -    linecnt 2 0
430672 -        ..
430673 -       Scroll up to avoid writing over HTML just entered, and save
430674 -       location to come back after determining the number of lines that
430675 -       need to be added to the record so that opening the last anchor
430676 -       will display properly at the top of the browser.  Find the last
430677 -       anchor by zoning to look for double dots, and do a reverse
430678 -       search to put the last anchor in the record at the top of the
430679 -       screen.
430681 -        ..
430682 -       Move the cursor to that line and capture the line number in
430683 -       order to calcualte whether, and how many, lines need to be added
430684 -       to the record in order for the last anchor line to display in
430685 -       the browser.
430686 -
430688 -  ..
430689 - Line 2890, ref OF 5 E29F, -label nCn in 070405 about 40 lines below
430690 -
430691 -
430692 -    macro 21
430693 -    linecnt 3 0
430694 -    subcnt 3 @2
430695 -    -if @3 > 50 -goto nalttr
430696 -    setcnt 2 50
430697 -    subcnt 2 @3
430698 -    ins_text!<BR>!
430699 -    immed 2@r
430701 -        ..
430702 -       Return to the line where the search was launched below the
430703 -       ending HTML code.  Identify the line number of the record at
430704 -       this location and subtract the line number for the last anchor.
430705 -       If it is greater than 50, then jump ahead to enter ending HTML
430706 -       code for the file. ref SDS 0 TR6O
430708 -        ..
430709 -       If not, then subtract the current difference from 50 to
430710 -       calculate the number of additional lines needed at the bottom of
430711 -       the record so that the last anchor line will display properly,
430712 -       assuming 50 is the most that would be needed.  Create a BR
430713 -       command because PRE is now ended, and add that many lines with
430714 -       BR commands to the bottom of the record, so that the last anchor
430715 -       will display at or near the top of the browser.
430716 -
430718 -  ..
430719 - Line 2650, ref OF 5 7E9O, -label nalttr in 070405
430720 -
430721 -    bot
430722 -    down
430723 -    -label nalttr
430724 -    setcnt 0 0
430725 -    ins_text!</BODY>!
430726 -    up
430727 -    ins_text!</HTML>!
430728 -    -goto over
430729 -
430730 -       After entering HTML that ends font and pre commands, per above,
430731 -       ref SDS 0 0O6I, and after adding any additional lines needed for
430732 -       the last anchor in the record to display in a browser, per
430733 -       above, ref SDS 0 F88T, do a down to place the code that stops
430734 -       HTML processing on the last two lines of the file.  Then do the
430735 -       next step to complete processing.
430736 -
430737 -
430738 -
430739 -
430740 -
4308 -

SUBJECTS
Evaluation
Planning

4504 -
450501 -  ..
450502 - Document Conversion to HTML Modified According to W3C Standards
450503 -
450504 - Background
450505 -
450506 -    On 021102 shows background leading up this work. ref SDS 42 I74J
450507 -
450508 -    On 021106 received template files from Gary Johnson for guidance on
450509 -    improving HTML code in macros 0122 and 070405. ref SDS 45 XY4E
450511 -  ..
450512 - Today, I am going to try changing the code to produce work product
450513 - that uses formatting in the template files Gary submitted on 021106.
450514 -
450515 -         Original letter, ref DRP 5 PRQV, is dated 021028,
450516 -         ref DIP 1 0001,
450517 -
450518 -         Corrected version has Gary's comments. ref DRP 5 0952
450519 -
450520 -
450521 -
450522 -
450523 -
4506 -

SUBJECTS
Documents 0122 Doc Type

4803 -
480401 -  ..
480402 - DocType Declaration Missing from Documents
480403 -
480404 - Follow up ref SDS 45 XY93.
480405 -
480406 - Gary proposes adding a doc type to all files....
480407 -
480408 -     Documents, ref DRP 5 0952,
480410 -  ..
480411 - For now I am going to skip this until there is more information on
480412 - analysis in the record on 021102. ref SDS 42 JL5O
480413 -
480414 -
480415 -
480416 -
4805 -

SUBJECTS
Letterhead

4903 -
490401 -  ..
490402 - Letterhead for Documents Converted to HTML
490403 -
490404 - Follow up ref SDS 24 0636.
490405 -
490406 - The code we now produce looks like this...
490407 -
490408 -     <A NAME="xx"></A><A NAME="0000"></A></font><font FACE="Times Roman" SIZE=4>
490409 -     <center></font><font FACE="Times Roman" SIZE=3 COLOR="red"><B>
490410 -     THE WELCH COMPANY<BR>
490411 -     440 Davis Court #1602<BR>
490412 -     San Francisco, CA  94111-2496<BR>
490413 -     415 781 5700<BR></B>
490414 -     </center>
490415 -     </font><font FACE="Times Roman" SIZE=4 COLOR="black">
490417 -  ..
490418 - The code Gary proposes looks like this...
490419 -
490420 -     <A NAME="xx"></A><A NAME="0000"></A>
490421 -     <center><font FACE="Times Roman" SIZE=3 COLOR="red"><B>
490422 -     THE WELCH COMPANY<BR>
490423 -     440 Davis Court #1602<BR>
490424 -     San Francisco, CA  94111-2496<BR>
490425 -     415 781 5700<BR></B>
490426 -
490427 - ...in his template file. ref DRP 5 H64H
490429 -  ..
490430 - Line 1290, ref OF 2 1392, -label keepit in 0122
490431 -
490432 -    ins_text!</A><A NAME="0000"></A><font FACE="Times Roman" SIZE=4 !
490433 -
490434 -       Eliminated the xx anchor, since it seems to be an original idea
490435 -       that has never found an application.
490437 -        ..
490438 -       Eliminated /font
490439 -
490441 -  ..
490442 - Line 2410, ref OF 2 1T3K, -label cadres in 0122
490443 -
490444 -    z 1 6
490445 -    l /<! sen/
490446 -    -if @0 > 0 -goto memor
490447 -
490448 -       Look for the Sender address field starting at the of the file,
490449 -       and when they have been processed, go do the memos headings.
490450 -
490452 -  ..
490453 - Line 2440, ref OF 2 2312, -label cnADR in 0122
490454 -
490455 -      ins_text!<center></font><font FACE="Times Roman" SIZE=3 COLOR="red">!
490456 -    ins_text!</font><center><font FACE="Times Roman" SIZE=3 COLOR="red">!
490457 -
490458 -       Try unplugging the code with /font after center, and puttting
490459 -       it in front.
490461 -        ..
490462 -       This is an offending code because it has /font, per Gary's
490463 -       example showing this is incorrect, see above. ref SDS 0 JT4G
490465 -        ..
490466 -       Removing /font is not a problem for the first letterhead at the
490467 -       top of a file, but this is a subroutine to process all
490468 -       letterheads that may be in a file, as occurs with attachments.
490469 -       In the case of attachments, there must be a pending font command
490470 -       to turn off, since otherwise we would not need a new font
490471 -       command.
490473 -        ..
490474 -       Thus, I am reluctant to revise this code.  We might rig
490475 -       something that treats the first letterehad differently by
490476 -       removing the /font.
490478 -        ..
490479 -       The criteria used for developing HTML conversion
490480 -       formatting is in...
490481 -
490482 -                        HTML
490483 -                        The Definitive Guide
490484 -                        2nd Edition
490485 -                        O'Reilly & Associates, Inc.
490486 -                        Copyright 1997
490487 -                        IBSN:  1-56592-235-2
490489 -           ..
490490 -          Page 43
490492 -           ..
490493 -          3.3.6  Omitting Tags
490494 -
490495 -             You often see documents in which the author seemingly has
490496 -             forgotten to include many endings in apparent violation of
490497 -             the HTML standards.  But your browser doesn't complain, and
490498 -             the documents displays [sic] just fine.  What gives?  The
490499 -             HTML standard lets you omit certain tags or thier endings for
490500 -             clarity and ease of prepation.  The HTML standard writers
490501 -             didn't intend the language to be tedious.
490503 -              ..
490504 -             Rather, the HTML standard lets you omit a starting tag or
490505 -             ending tag whenever it can be unambiguously inferred by
490506 -             the surrounding context.  Many browsers make good guesses
490507 -             when confronted with missing tags.
490509 -        ..
490510 -       The p tag is cited as an example where an ending tag is rarealy
490511 -       used by anyone, but the author goes on to say...
490512 -
490513 -             When in doubt add the ending tag:  it'll make life easier
490514 -             for yourself, the browser, and anyone else who might need
490515 -             to modify your document in the future.
490517 -           ..
490518 -          3.3.6  Ignored or Redundant Tags
490519 -
490520 -             Browsers sometimes ignore tags.  This usually happens with
490521 -             redundant tags whose effects merely cancel or substitute
490522 -             for themselves.
490524 -           ..
490525 -          Page 44
490526 -
490527 -             Browsers habitually forge ahead and make some sense
490528 -             of a document, no matter how badly formed and error ridden
490529 -             it may be.  This isn't just a tactic to overcome errors,
490530 -             it is also an important strategy for extensibility.
490531 -
490533 -  ..
490534 - Line 2510, ref OF 2 3198, -label prec in 0122
490535 -
490536 -      ins_text!</center !
490537 -    ins_text!</font></center !
490538 -    immed a
490539 -    up
490540 -      ins_text!</font><font FACE="Times Roman" SIZE=4 COLOR="black" !
490541 -    ins_text!<font FACE="Times Roman" SIZE=4 COLOR="black" !
490542 -
490543 -       Tried unplugging /center and adding /font /center, then do a
490544 -       font op.
490545 -
490546 -
490547 -
490548 -
490549 -
490550 -
490551 -
4906 -

SUBJECTS
Date and Doc ID Table

5003 -
500401 -  ..
500402 - Date and Doc ID Table Corrected Font and TD Tag Problem
500403 -
500404 - Line 3420, ref OF 2 5884, -label ddaimd about 30 lines below
500405 -
500406 -    immed 3a
500407 -    up
500408 -     ins_text!<p align="left">!
500409 -     up
500411 -        ..
500412 -       Turns out we do not need para align command now that we are
500413 -       using center for the letterhead, so can eliminate and adjust
500414 -       line spacing.
500416 -     ..
500417 -    ins_text!<tr> <td WIDTH="65%" VALIGN="TOP">!
500418 -    up
500419 -    ins_text!<font FACE="Times Roman" SIZE=4 COLOR="black">!
500420 -
500421 -                    Date and right justify cell
500422 -    up 2
500423 -    ins_text!</font></td><td WIDTH="35%" VALIGN="TOP"><p align="right">!
500424 -    up
500425 -    ins_text!<font FACE="Times Roman" SIZE=4 COLOR="black">!
500426 -
500427 -                   Table ending code - skip over date field to blank line
500428 -                   below
500429 -     e.
500430 -    eol
500431 -    ins_text!</font>!
500432 -    immed a
500433 -       ..
500434 -       Tried changing this stuff to work better.  Think have
500435 -       added missing /td and /font commands.
500436 -
500437 -
500438 -
500439 -
500440 -
500441 -
500442 -
500443 -
5005 -

SUBJECTS
Entities < & > Code for Display in Browser

5103 -
510401 -  ..
510402 - Entities in Documents Difficult to Convert to HTML
510403 -
510404 - Today, we have added code to handle conversion of entities in SDS
510405 - records, per above. ref SDS 0 5S4J
510407 -      ..
510408 -     Not sure how this would work in a document file that is created
510409 -     with HTML code.  How does the conversion process know the
510410 -     difference between entities that are part of the message, and HTML
510411 -     code?  We could create a function key, say F1 F7, to process a
510412 -     line of text.  The User could enter the actual characters, then
510413 -     visually recognizing something that is not HTML formatting, call
510414 -     F1 F7 that performs conversion to entities, but this leaves the
510415 -     text with entity commands which are hard to comprehend, so it is
510416 -     not a good solution.
510418 -      ..
510419 -     Maybe Gary can comment.
510420 -
510421 -        [On 021116 Gary submits ideas, says general solution is
510422 -        difficult. ref SDS 48 0001
510423 -
510424 -
510425 -
510426 -
510427 -
510428 -
510429 -
510430 -
510431 -
5105 -

SUBJECTS
Address

520301 -  ..
520302 - Address Memo Format Enhanced Process to Integrate with To From
520303 -
520304 - The code that processes From and To address fields for memoradums is
520305 - not processing the To field correctly.  I think the problem is that we
520306 - have added an "<address> field, and this is jumping past the To field,
520307 - and just doing the From field.
520309 -  ..
520310 - There are several ways to handle this.
520311 -
520312 -      1.  When a Memorandum is used, don't use address formatting.
520313 -
520314 -      2.  When a To field is used don't use address formatting.
520315 -
520316 -      3.  Use address and place a blank line between address and To,
520317 -          this may trick the code to creating the correct formatting.
520318 -          ..
520319 -      4.  Modify address processing to initially look for To, and
520320 -          if there, do To, and not standard address process.
520322 -  ..
520323 - Decided to use option 4 because correspondence comes in with
520324 - address already there, and so it is easy to forget that something
520325 - special needs to be done for To.
520327 -  ..
520328 - Line 1850, ref OF 2 G65L, -label adrslp in 0122
520329 -
520330 -    fr/To:/
520331 -    -if @0 = 0 -goto adrTo
520332 -    up
520333 -    fr/TO:/
520334 -    -if @0 = 1 -goto adrTO
520335 -
520336 -       In Address processing, look for To and TO and if within first
520337 -       three positions, process those formats.
520338 -
520339 -
5204 -

SUBJECTS
Subjects, Regarding, Re

530301 -  ..
530302 - Subject Table COLS Parm Seems to Reduce Flexibility of Window Size
530303 -
530304 - Having difficulty getting the Subject field to display.  An example is
530305 - the letter in the record on 021028. ref SDS 41 253O
530306 -
530307 - Line 3460, ref OF 2 AZ7L, -label tfsub in 0122
530308 -
530309 -      ins_text!<table BORDER="0" CELLSPACING="0" CELLPADDING="0" CELLS="2" WIDTH="85%">!
530310 -      ins_text!<table BORDER="0" CELLSPACING="0" CELLPADDING="0" COLS="2" WIDTH="85%">!
530311 -    ins_text!<table BORDER="0" CELLSPACING="0" CELLPADDING="0" WIDTH="85%">!
530312 -
530313 -                 Subject
530314 -
530315 -    -if @138 = 3 ins_text!<tr> <td WIDTH="15%" VALIGN="bottom">!
530317 -     ..
530318 -                 Description
530320 -     ..
530321 -    ins_text!</td><td VALIGN="bottom" ALIGN="left">!
530322 -
530323 -       Gary pointed out that the CELLS="2" should be COLS="2" - so I
530324 -       made this change to COLS="2" with a max line width 85% of screen
530325 -       width, since that is standard for correspondence.  COLS=2 seems
530326 -       correct because we have two columns.  We, also, have a cols=2
530327 -       parm in the table command for the date and ID field and it seems
530328 -       to work. ref OF 2 DN3J
530330 -        ..
530331 -       I have been experimenting with Netscape 7.0, (reported later on
530332 -       021114, ref SDS 47 0001), and when I made the change to COLS=2
530333 -       the configuration displayed correctly using Netscape 7.0.  So, I
530334 -       assumed, without really thinking about it, nor testing, that
530335 -       this change would display the same way with IE and Netscape
530336 -       4.73, and changed all of the correspondence through last July to
530337 -       use this new configuration, thinking it would be helpful for
530338 -       Gary on the assumption that correction to aid Netscape 7.0 would
530339 -       work similarly in Opera.
530341 -        ..
530342 -       Later, I was working on something else and noticed that IE was
530343 -       displaying the 2nd part of the Subject field right justified, as
530344 -       if we had an p align="right" command, when in fact the command
530345 -       is p align="left", per above. ref SDS 0 S56L  Display with
530346 -       "right" instead of "left" alignment not only ignores an express
530347 -       command, it creates a big gap between the field title Subject
530348 -       and the content, which is not a professional rendering.
530350 -        ..
530351 -       I checked the same letter in Netscape 4.73 and it showed the
530352 -       same thing.  I checked the letter with Netscape 7.0 and it
530353 -       showed the correct display, indicating that Netscape 7.0 handles
530354 -       things differently from both IE and Netscape 4.73.  This meant
530355 -       that the hundred or so letters had been changed for Gary were
530356 -       all incorrect, unless Opera works the same way as Netscape and
530357 -       differently from IE and Netscape 4.73.
530358 -
530359 -           [On 021115 need consistent display by browser, decided to
530360 -           wait on using Netscape 7.0 browser until it reaches level of
530361 -           performance for Netscape 4.73 and IE 6.0. ref SDS 47 0001
530363 -            ..
530364 -           [On 021115 other display problems with Netscape 7.0.
530365 -           ref SDS 47 XM4J
530367 -        ..
530368 -       After investing 3 or 4 hours to get IE and Netscape 4.73 to
530369 -       display the subject field correctly in the way that Netscape 7.0
530370 -       does, by a process of elimination I tried removing the cols=2
530371 -       command.  This produced a correct and consistent rendering for
530372 -       the Subject field by all three browsers.
530374 -        ..
530375 -       This record reveals a number of false positive connections that
530376 -       lead the mind down the wrong path, per review of Landauer's
530377 -       theory of meaning and learning derived from associations,
530378 -       reviewed on 960324. ref SDS 4 8844  The record here shows how
530379 -       easy it is to become convinced that an error seems correct....
530380 -
530381 -          1.  Since we have two fields, cols="2" seems like the right
530382 -              choice.
530384 -               ..
530385 -          2.  Since cols="2" is used for the date and ID field, that
530386 -              further indicates it is the right configuration for the
530387 -              subject field.
530389 -               ..
530390 -          3.  IE and Netscape 4.73 displayed the date and ID field
530391 -              properly because, in that case, the doc ID element should
530392 -              be right justified.  It was assumed the p align="right"
530393 -              command caused this correct alignment, since there is not
530394 -              an evident reason to link the cols command with
530395 -              alignment, rather than the alignment command.
530396 -              ..
530397 -              The fact that cells="2" was an incorrect command
530398 -              further misled the mind from several years of correct
530399 -              rendering. Evidently IE and Netscape 4.73 ignored this,
530400 -              so there was no history of evidence to suggest that
530401 -              cols="2" impacts alignment.
530403 -        ..
530404 -       One evident advantage of eliminating cols=2 in table commands
530405 -       for both the date and doc ID, and the subject fields is
530406 -       increased flexibility in sizing the window by a reader.  With a
530407 -       cols=2 command, when the window is reduced in size, the strings
530408 -       in the 2nd column bunch up in a vertical stack, which corrupts
530409 -       that formatting.  Without cols=2, the window has a much broader
530410 -       range of reducing the width before the same stacking effect
530411 -       occurs.
530413 -        ..
530414 -       It would be interesting to see analysis comparing explanations
530415 -       of HTML standards with experience based on applying actual code
530416 -       configurations.
530418 -        ..
530419 -       Once cols="2" was eliminated from the table command, per above,
530420 -       ref SDS 0 654F, the Subject field displays correctly in all
530421 -       three browsers, and adjusts in an acceptable way when the window
530422 -       size is increased or decreased.
530424 -        ..
530425 -       Spent another 20 minutes using the doc log to change several
530426 -       hundred documents to the correct configuration.  Hope this will
530427 -       work okay for Opera.
530428 -
530429 -
530430 -
5305 -

SUBJECTS
Subjects, Regarding, Re

5403 -
540401 -  ..
540402 - Post Script Format No Spaces Below Signature Block
540403 -
540404 - Line 2670, ref OF 2 1N8O, -label PSlp in 0122
540405 -
540406 - If there is a post script field, we want it positioned directly below
540407 - the signature block separated only by an anchor.
540409 -  ..
540410 - It has to start out at least three three (3) lines below the bottom of
540411 - the signature block in order for processing to occur correctly for the
540412 - signature block, reported on 021013. ref SDS 35 07CH
540414 -  ..
540415 - One thing we can do is verify the spacing in advance of processing
540416 - the file and put the correct in place.
540417 -
540418 -
540419 -
540420 -
540421 -
540422 -
540423 -
540424 -
540425 -
540426 -
5405 -

SUBJECTS
Para P Commands
BR BR Command Adopted to Replace P
BR Command

5905 -
590601 -  ..
590602 - P Command Switch to BR BR to Avoid Standards Issues
590603 -
590604 - Follow up ref SDS 42 P15I.
590605 -
590606 - A number of issues seem to arise using the p command that may be
590607 - avoided by substituting br br, per analysis on 021102. ref SDS 42 PP5M
590609 -  ..
590610 - Line 2690, ref OF 2 2260, -label paraP in 0122
590611 -
590612 -      -if @108 = 1 ins_text!<P>!
590613 -
590614 -    -if @108 = 1 ins_text!<br><br>!
590615 -
590616 -       To try and reduce problems with p command changed to br br.
590618 -  ..
590619 - Line 2920, ref OF 2 R23L, -label caanc in 0122 about 120 lines below
590620 -
590621 -    setcnt 0 0
590622 -    loc_cur 0 1
590623 -    fr/<br><br>/
590624 -    -if @0 = 0 immed d
590625 -
590626 -       Added code to look for new string br br as para separator where
590627 -       a redundant link is entered for addressability.
590629 -  ..
590630 - This seems to work just as well.
590631 -
590632 -     [On 021114 Gary reports SDS records and documents now display in
590633 -     Opera, indicating closer compliance with HTML standards.
590634 -     ref SDS 46 0001
590635 -
590637 -  ..
590638 - Line 480, ref OF 3 2155, -label bX017F06  in 04707
590639 -
590640 -    This code recognizes that the F6 has been pressed to enter a P
590641 -    command.  Today, we change this to enter a br br command.  This was
590642 -    accomplished by changing the entire scheme.  At this location,
590643 -    merely set a flag to establish that F6 has been pressed to enter
590644 -    either a BR or a br br command.
590646 -     ..
590647 -    setcnt 2 10
590648 -    -goto HTm1a
590649 -
590650 -         Flag says F6 is pressed, then go to standard setup.
590652 -  ..
590653 - Line 190, ref OF 3 VP4G, -label HTm1a in 04707
590654 -
590655 -    -label HTm1a
590656 -    setcnt 169 0
590657 -
590658 -        Not sure why this is needed for the BR and br br commands, but
590659 -        will leave it for now.
590660 -
590661 -    -label HTm2
590662 -    errorcnt 0 0
590663 -    savescr 14
590664 -    -if @2 = 10 -goto bRBR
590665 -
590666 -        Set up for fr command and to restore the screen.  For BR and
590667 -        br br when F6 is pressed, go do that code.
590668 -
590669 -
590671 -  ..
590672 - Line 360, ref OF 3 U97L, -label bRBR in 04707
590674 -  ..
590675 - Changed code to enter enter <br><br> instead of <P> command,
590676 - per above, ref SDS 0 ZE4F, and change BR command so it will
590677 - overwrite br br to enter a single br command.
590679 -     ..
590680 -    -label bRBR
590681 -    z 1 250
590682 -    of 0
590683 -    loc_cur 0 1
590684 -    fr !<BR!
590685 -    -if @0 = 0 -goto br2r
590686 -
590687 -        The code will look to see if there is a BR command already on
590688 -        the line, and if so, it will go an change BR to br br, per
590689 -        below. ref SDS 0 SB4O
590691 -     ..
590692 -    fr !<br!
590693 -    -if @0 = 1 rel_cur 0 -3
590694 -
590695 -        The code did not find a BR so look for br, which means it is a
590696 -        br br or a P command operation.  If br is encountered, then
590697 -        move the cursor back to the beginning column for the command,
590698 -        ready to enter a new command.
590700 -     ..
590701 -    -if @0 > 1 eol
590702 -    ins_text!<BR>     !
590703 -    -goto gs14
590704 -
590705 -        If the code did not find br, the move the cursor to the end of
590706 -        the line, and enter a BR command; and then go and restore the
590707 -        file and the cursor to the initial position and exit from the
590708 -        operation, leaving the profile macro 04707 in memory for the
590709 -        next command.
590710 -
590712 -  ..
590713 - Line 380, ref OF 3 WA9G, -label br2r in 04707
590714 -
590715 -    -label br2r
590716 -    rel_cur 0 -3
590717 -    ins_text!<br><br>!
590718 -    -goto gs14
590719 -
590720 -        The code found that BR is at the end of the line, per above,
590721 -        ref SDS 0 DF7M, so replace this with a P command, which we are
590722 -        using br br.
590723 -
590724 -
590725 -
590726 -
590727 -
590728 -
5908 -

SUBJECTS
HTML End

6003 -
600401 -  ..
600402 - End Document Formatting
600403 -
600404 - Line 1310, ref OF 2 3591, -label keepit
600405 -
600406 -     ins_text!</font></body>!
600407 -     up
600408 -     ins_text!</HTML>!
600409 -
600410 -        Added /font to end formatting.
600411 -
600412 -
600413 -
600414 -
600415 -
600416 -
600417 -
6005 -

SUBJECTS
Save Original Position and Restore After Converserion to HTML for Onl

6103 -
610401 -  ..
610402 - Restore Original Position in Document After Conversion
610403 -
610404 - We need a way to restore the original position of a file that is
610405 - converted to HTML, so that further editing can occur.  This is not an
610406 - issue with SDS records, because nothing is added.
610408 -  ..
610409 - Documents are changed by 0122 by adding anchors to most every para for
610410 - increased addressability.  Actually, we could do this with F2 save
610411 - function, as occurs with saving SDS records, but up until now, save in
610412 - documents has simply written the file to the disk, whereas saving SDS
610413 - records has always required a lot of processing.  So addressability
610414 - added to SDS records was done along with all the other processing that
610415 - occurs with "save" to take advantage of common code that keeps up the
610416 - speed and makes everything seem transparent.  This means anchors are
610417 - not added to SDS records with 070405 conversion to HTML.  However,
610418 - when a document is converted to HTML anchors are added by 0122 for
610419 - addressability.  Therefore when a document comes back to the User, the
610420 - relative position can be significantly different, at least on the
610421 - first pass, and when subsequent editing occurs.
610423 -  ..
610424 - In any case, we need a way to restore the User to the original
610425 - perspective in the document.  Press a key, something happens, the
610426 - user expects to see the same thing that was seen before pressing the
610427 - key, otherwise confusion occurs.
610428 -
610430 -  ..
610431 - Line 150, ref OF 2 758M, -label saHTML in 0122 about 70 lines below
610432 -
610433 -    Want to revise method of restoring original file; instead of
610434 -    reading a string and finding it later, will save a string, and
610435 -    enter a unique string, then find that string when the revised file
610436 -    is read back, and replace it with the original string.  This will
610437 -    permit launching the conversion operation on a blank line.
610439 -     ..
610440 -    setcura 236 0
610441 -    loc_cur 0 60
610442 -    chrcnt 235 0
610443 -    setgbl 235 @235
610444 -
610445 -       Get relative location of cursor to restore at end of process. We
610446 -       cannot use standard mark and return because everything may be
610447 -       moved when the file is restored after anchors are added. sectura
610448 -       236 puts the y coordinate in counter 237 which is what we need,
610449 -       since the x coordinate will be constant.  Put the flag character
610450 -       in column 60, so it will not be an anchor, since the first that
610451 -       occurs in the dose session is to add anchors.  If code
610452 -       encounters the flag character, then it wil not read as an
610453 -       anchor, and this would cause the process to fail, putting the
610454 -       flag character in column 60 avoids this because anchors are
610455 -       entered in columns 1 40 or something like that.
610457 -     ..
610458 -    ic 30
610459 -
610460 -       Enter a control character that does not show up on
610461 -       ordinary document creation. we are using *
610463 -  ..
610464 - Line 190, ref OF 2 H33I, -label saHTML in 0122 about 90 lines below
610465 -
610466 -    e c:\sd\03\0122
610467 -    line
610468 -    loc_cur 4 3
610469 -    getgbl 235 235
610470 -    ic @235
610471 -    save
610472 -
610473 -       Open the conversion macro 0122 and store the actual character to
610474 -       swap things out under the dose session.  Save 0122, so when it
610475 -       is opened under dose session, the code can get that char to
610476 -       restore it to the original location after anchors have been
610477 -       added so the correct file can be processed with HTML and moved
610478 -       to the transfer directory, per below, ref SDS 0 4H5F, but before
610479 -       saving the modified with anchors, so that the control char can
610480 -       be used to restore the initial condition to the User, then it is
610481 -       replaced in this next step....
610482 -
610484 -  ..
610485 - Line 240, ref OF 2 4Z4N, -label saHTML in 0122 about 130 lines below
610486 -
610487 -    z 1 80
610488 -    s !*!
610489 -    fr !*!
610490 -    rel_cur 0 -1
610491 -    ic @235
610492 -    rel_cur 0 -1
610493 -    save
610494 -
610495 -       After the document file has been converted to HTML and we read
610496 -       in the revised file with HTML, restore the original character
610497 -       and save the file.
610498 -
610499 -
610501 -  ..
610502 - Line 360, ref OF 4 HR7H, -label nnwo in 0608 about 20 lines below
610503 -
610504 -    @c:\sd\03\0122
610505 -
610506 -       0122 is called at the primary level.  We need to know this.
610507 -
610509 -  ..
610510 - Line 480, ref OF 2 UL7G, -label anLP1 in 0122
610511 -
610512 -    Anchors are added to all para for addressability at this location.
610513 -
610514 -
610516 -  ..
610517 - Line 750, ref OF 2 SE5L, -label saVV in 0122
610518 -
610519 -    save
610520 -    top
610521 -
610522 -       We are in the dose session, and the file has been saved with
610523 -       anchors added and with the temp char marker entered wherever it
610524 -       happened to land.
610526 -        ..
610527 -       We now look for replace the temp char with the actual character
610528 -       so that when the file is saved to the transfer directory and to
610529 -       the local Internet data base, it will have the correct char.
610531 -     ..
610532 -    z 45 90
610533 -    s !*!
610534 -    fr !*!
610535 -    rel_cur 0 -1
610536 -
610537 -       We are on the location.
610539 -     ..
610540 -    markcur 10
610541 -    e c:\sd\03\0122
610542 -
610543 -       Mark the target file and open 0122 processor to get the actual
610544 -       char saved in the step above. ref SDS 0 R44G
610546 -     ..
610547 -    line
610548 -    loc_cur 4 3
610549 -    chrcnt 235 0
610550 -    setgbl 235 @235
610551 -    ecur 10
610552 -
610553 -       Get the char and return to the target file, restore the
610554 -       char
610555 -
610556 -
610557 -
610558 -
610559 -
610560 -
610561 -
610562 -
610563 -
610564 -
6106 -
Distribution. . . . See "CONTACTS"