Hi Danny,

I will pass on your comments to Lew.  

It would certainly sound logical for the underlined DP to have an underline 
output  but even in XHTML the <u> is deprecated (but still available) in favour 
of using a style. So perhaps if there are application and scenarios using <i> 
it might be better to leave them alone. 

Actually on a more philosophical note, you rarely see underlines except for 
hyperlinked data, and the most common text decorations are either bold or 
italic. 

In terms of the expansion of the DP memo field, it might be a little too much 
for this version. I know Lew was joyous about dropping the bare <p> to 
represent two consecutive lines breaks as he has taken some things to the limit 
and saving that extra special character was a benefit to him.

Regards
Brian

  ----- Original Message ----- 
  From: MEIRTE Danny 
  To: [email protected] 
  Sent: Wednesday, June 11, 2008 1:33 AM
  Subject: [Dataperf] new DP 2.Y


  Dear Lew (and all dp users),



  Nice to hear about a DP2.6 Y version on its way … (and even more for a DPWin 
version)!



  The problem with the underline and italics is that nowadays we actually need 
both. But if we have to choose, I would prefer italics (in biology and 
scientific references it is quite important).

  Maybe it is the moment to ADD italics as a third font-style (additional to 
bold <F6> in text edit, underline <F8> in text edit). I know this would imply a 
special color scheme to visualise them (and the range of combinations possible 
would increase from 5 to 9 …). Maybe <F4> in text edit can be used for keyboard 
input of italics (the WP4.2 tags are <178> <179>).

  If this is too complicated, one could also leave the option to translate the 
underline marks into italics or not at a higher level (at database STR or dp 
startup level). (the same way as it has been solved in DPSpool…)



  Suggestion for default settings: [ctrl-F3, 2]

  1 - Menu Color: B H

  2 - Editing Color: H A

  3 - Bold Color: B P

  4 - Underline Color: C O

  5 - Italics Color: B D

  6 - Bold & Underline: C P

  7 - Bold & Italics: B L

  8 - Italics & Underline: C D

  9 - Bold, Italics & Underline: C L



  Note: The only missing “font-style/variant” is smallcaps. But as you probably 
know this is well supported in RTF but is not so ‘regular’ in HTML (only 
through styles… the ‘shortest’ HTML-output looks like: <span 
style=font-variant:small-caps> </span> ).

  (the total number of color combinations will be 17!, although SmallCaps is 
mostly a stand-alone or combined only with Bold; the <F2> in text edit can be 
used fpr keyboard input of smallcaps) (no WP4.2 tags known to me, but <144> 
<145> are used in my Printcodes)



  Remark: 

  I do not use DP2.6X because I am working in a special multilingual codepage, 
where the translation from DOS (CP 437) to ISO Latin 1 is not appropriate. 

  Please note that CP 437 (US), CP 850 (Multilingual) and CP 852 (Slavic) are 
quite different e.g. in character code 252: the n-superscript in CP 437, the 
2-superscript in CP 850 and r-hacek in CP 852.


  In Europe the euro symbol € is needed as currency output char(238) in CP 437, 
so CP850 changed into CP858, by dropping the dotless-i; CP 852 is still without 
euro symbol… (char(238) is t-cedilla). The ISO 8859-1 (Latin-1) does not hold 
the euro symbol, it is adapted to the ISO 8859-15 or Latin 9 that holds the 
euro symbol as char(164)… 

  In HTML pages the euro symbol can be encoded using SGML character references 
as &euro; or using Unicode &#8364;



  I would also suggest to make the ;;W translation of each character 
adaptable/programmable by the user: I would hope for a similar interface as the 
Printer Map interface [Ctrl-F3; 6] but allowing 4 bytes in the output (this is 
the maximum length for a character in UTF-8), or 9 bytes for HTML-code using 
Unicode [e.g. male symbol: &#9794; and female &#9792;), or 10 bytes for SGML 
character references (longest I found 10 bytes &thetasym; same as &#977; in ISO 
Latin 1 the longest are 8 characters long e.g. &plusmn; same as &#177;)

  Of course, it would be even more flexible to have the ;;U (for UTF-8) and ;;W 
(for HTML-code) as separate output format suffixes, each using their own 
translation table…



  Please note that the Microsoft CP 1252 ("Windows Latin-1") is not the same as 
the Official ISO Latin-1. In real ISO Latin-1, character codes in the range 
127-159 are undefined. Microsoft has assigned these undefined codes to glyphs 
listed on their codepage CP 1252.

  It is recommended to use Unicode UTF-8 for encoding symbols that are not in 
the 7-bit ASCII set, as it is the same across all platforms and operating 
systems that support it.

  For XML UTF-8 is the standard encoding!



  __________________________________________

  Just as it might be a lucky moment for us all…



  Some (old) wishes (suggestions are for DP2.6, I use DP2.6f for the moment, 
but I checked my problems using DP2.6X):

  1) Is it possible to fix the bug that during reports the “Please Wait” and 
“Network Busy” messages would restore the cursor position on the screen 
properly? 

  2) A second problem (memory allocation bug?) to fix: when using a calculated 
(virtual) field for retrieving a record through a record link using [F5], the 
first multiline alphanumeric memo type field (not even used in the calculation) 
gets corrupted… [Probably I have to document this in more detail].

  3) Maybe the record counter during search (now resetting at 65k) could be 
adapted.

  4) suggestion: show fieldnames (as in search) in the Edit/Create import list 
[Ctrl-F5, 7]

  5) suggestion: allow inverse indexing (showing highest first) in a panellink 
and in subreports as it is possible in the main report (sort direction: 
backward) and probably also in a Related Records Window (could be indicated by 
using a negative value for the link description in the STE format)

  6) hoping for some trigonomial functions, the most needed are: sin[number] 
for sinus with number in radians and atn[number] for arctangent in radians 

  7) the solution for the complex calculations in my database [now using up 
many panellinks]: a lookup through an existing index (indicated by a 
panelnumber and a filenumber) not with a fixed panellink but using variables,…

  suggestion: lookup[panelnumber;indexnumber;indexvalue;n] (a negative 
indexnumber would inverse the query – see (5)-); n indicates the field to 
return (similar to user.field[n]); an other way could be 
FIND[P1F4P2F3;indexvalue] where P2 is the panelnumber P1F4 is the panellink 
using indexnumber and F3 corresponds with n) –the latter format is not quite 
DP-like-

  8) trespassing the 78 character limit for SUBFIELD and SUBSTRING (at least 
when used in report variables).

  9) for SUBFIELD: syntax for using linefeed, bold, underline (and italic!) 
codes for parsing

  10) a function to be used inside a formula allowing to put text to bold, 
underline (and italic). suggestion: functions cat.b[ ], cat.u[ ] (and cat.i[ ]) 
same as cat.c[ ] but resulting text will be bold, underlined resp. in italics. 
And why not: a cat.n[ ] to strip off all these codes and leaving only linefeeds…

  11) for template search on a memo field: a way to look for linefeed, bold, 
underlined, (italics,) … [might be the same solution as for (9)!]



  I hope I contributed something…



  Danny





  Danny Meirte

  curator Herpetology

  Royal Museum for Central Africa

  Department of African Zoology

  Vertebrate Section

  Leuvensesteenweg 13

  3080 Tervuren, Belgium

  Tel.: +32 2 769 56 29 Fax : +32 2 769 56 42 or +32 2 767 02





  ###########################################

  This message has been scanned by ICT - Africa Museum

  ________________________________________ 

  10/6/2008 - Filtered through antispam by ICT



------------------------------------------------------------------------------


  _______________________________________________
  Dataperf mailing list
  [email protected]
  http://lists.dataperfect.nl/mailman/listinfo/dataperf
_______________________________________________
Dataperf mailing list
[email protected]
http://lists.dataperfect.nl/mailman/listinfo/dataperf

Reply via email to