[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #11 from Commit Notification --- Eike Rathke committed a patch related to this issue. It has been pushed to "master": https://git.libreoffice.org/help/commit/59861677cc4ca59020718033f675b54f49fc0b74 Size in characters, not bytes; tdf#153574 follow-up -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574
--- Comment #10 from Eike Rathke ---
(In reply to ady from comment #9)
> * In this context, is "1024 characters" the same as "1024 bytes"?
No. It should be 1024 characters (that detail slipped my attention when I took
a look), or to be precise 1024 UTF-16 code points (though users probably
wouldn't know what that is), which can be less than 1024 characters if
surrogates are used.
> For the Interpreter, it now says 256MB in size. Is the "256MB" accurate?
Well it's max signed int32 / 8 so 2147483647/8 that is 268435455
(268435455.875) so 255.99046326MiB ...
> =REPT("1234567890"; 10^8)
> ...already generates Err:513.
Because it would result in 10 (1*"1234567890") characters.
> I don't know how that amount of characters
> relates to "256MB" in this context.
It's greater than 256M characters.
> * Is this 256MB really the threshold?
Yes.
I'll change the slightly confusing bytes vs characters and 256MB (it's not
bytes either) wording.
--
You are receiving this mail because:
You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574
--- Comment #9 from ady ---
(In reply to Eike Rathke from comment #6)
> Err:513 btw the 64KB string length limit in the interpreter is now 256MB,
> and "Compiler: an identifier in the formula exceeds 64 KB in size" is wrong,
> the length limit for one identifier (function name, ...) is 1024 characters.
The new help file says instead (not counting for indentation here):
Compiler: an identifier in the formula exceeds 1024 bytes in size.
Interpreter: a result of a string operation exceeds 256 MB in
size.
* In this context, is "1024 characters" the same as "1024 bytes"?
For the Interpreter, it now says 256MB in size. Is the "256MB" accurate? For
instance:
=REPT("1234567890"; 10^8)
...already generates Err:513. I don't know how that amount of characters
relates to "256MB" in this context.
* Is this 256MB really the threshold?
--
You are receiving this mail because:
You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #8 from Commit Notification --- Olivier Hallot committed a patch related to this issue. It has been pushed to "master": https://git.libreoffice.org/help/commit/baa3ad9698dfc80375109478ff88d7929dfb62ea tdf#153574 Update error code Help page -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 Commit Notification changed: What|Removed |Added Whiteboard||target:7.6.0 -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #7 from Olivier Hallot --- first patch submitted https://git.libreoffice.org/help/commit/77b1a5a7118478c2b3a86fd43cfc68e336ac913c (bad bug number in commit msg) -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #6 from Eike Rathke --- I don't know who came up why with that "domain" thing.. Err:526 was an error that apparently could had been loaded from the old binary file format, which we don't support anymore, for some references thing related. It's gone and help for this can be removed. Err:525 #NAME? "no valid domain name" should be "no valid function name" or "unknown function name"; "incorrect decimal divider" is nonsense, should be "incorrect decimal separator" instead (e.g. entering 123,45 if the decimal separator is . dot and the function parameter separator is not , comma could lead to #NAME?) Err:504 "domain reference instead of cell reference" makes no sense, I also don't know what could had been meant instead. We usually set that IllegalParameter error when an actual argument's type does not match the expected type, or for unexpected parameters, but also on some other occasions. It's a kind of all-round-error if not other errors classified. Err:513 btw the 64KB string length limit in the interpreter is now 256MB, and "Compiler: an identifier in the formula exceeds 64 KB in size" is wrong, the length limit for one identifier (function name, ...) is 1024 characters. Err:514 "Sort operation attempted on too much numeric data (max. 10) or a calculation stack overflow" is wrong. It's only a compiler but more interpreter internal stack overflow error, not sort related. Err:538 (that currently has no description) is the sort maximum exceeded and the maximum is twice the maximum number of rows, so for 1048576 rows 2097152 entries. The same Err:538 is also used whenever a temporary matrix could not be allocated due to its size requirement (probably numbers not to be documented because easily misleading or subject to internal change: for 64-bit systems that's limited to 6GB which roughly fits 500 million numeric values (completely different number for strings); for 32-bit systems that's 1GB with ~85 million numeric values). Yes, Err:521 long error description was changed. -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #5 from ady --- (In reply to Olivier Hallot from comment #4) > What is your take? I don't have anything specific to comment regarding the use of the "domain" terminology. As I said, it is not a term that says much (or anything) to us, common users of spreadsheets. Now, since we are here already talking about: source/text/scalc/05/0214.xhp ...and there have been changes related to it lately, please allow me to add an additional request for correction: * In file: source/text/scalc/05/0214.xhp ...there is a line (not counting for indentation here): Internal syntax error (instead of Err:521 cell displays #NULL!) The above line includes the term "Internal syntax error". Based on: https://gerrit.libreoffice.org/c/core/+/150654 ...the term "Internal syntax error" is incorrect. Instead, it should be "No code or no intersection". -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 Olivier Hallot changed: What|Removed |Added CC||er...@redhat.com --- Comment #4 from Olivier Hallot --- (In reply to ady from comment #2) > Adding request for related minor corrections... (snip) > * In relation to Err:526: > Obsolete, > no longer used, but could come from old documents if the result is a formula > from a domain. > > ** The term "domain" is used in this help page in the errors: 504, 525 and > 526. As a user, I do not know what this term refers to in this context. This > is not good, considering that we are talking about a help page. > > Is the term "domain" (fully) equivalent to "(cell) range"? If it isn't, > where such expression is defined for this context? > Although I am intrigued by the term, I am not necessarily asking for an > answer to be written here (welcome anyway), but I'm asking for consistency > and clarity in the help files. If the term "cell range" is equivalent in > this case / context (which I do _not_ know), perhaps it would be better to > use a commonly-used term rather than some ambivalent less-used one, such as > "domain"? The "domain" nomenclature is sub-optimal for the end user. for error 526: It seems that "domain" is the current set of Calc functions. IMHO, the error is an "syntax error" from a function that may not exist anymore or has a different argument signature in the current set of valid Calc functions (including extension-provided functions). Comparing to Err:525, "no valid domain name" seems to indicate that the function is not on the current set of valid functions. Comparing to err:504, "a domain reference instead of cell reference" can mean a function name reference instead of a cell reference... What is your take? -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #3 from ady --- My apologies; for some reason in my prior comment I forgot to add the following item. For: text/scalc/05/0214.xhp * In relation to Err:525: An identifier could not be evaluated, for example, no valid reference, no valid domain name, no column/row label, no macro, incorrect decimal divider, add-in not found. ** If by "decimal divider" the intention is "decimal separator", please modify the term to the latter. -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 --- Comment #2 from ady --- Adding request for related minor corrections... For: text/scalc/05/0214.xhp * In relation to Err:525: invalid names (instead of Err:525 ... ** Capitalize "invalid names" > "Invalid names" in accordance to the rest of the items. * In relation to Err:540: Happens if a function that requires (re)loading of external sources is encountered and the user hasn't confirmed reloading of external sources yet ** Instead of "Happens if a function that requires...", it should simply be "A function that requires...", to be more consistent with the rest of the items. The sentence should also end with a stop/period/dot. * In relation to Err:526: Obsolete, no longer used, but could come from old documents if the result is a formula from a domain. ** The term "domain" is used in this help page in the errors: 504, 525 and 526. As a user, I do not know what this term refers to in this context. This is not good, considering that we are talking about a help page. Is the term "domain" (fully) equivalent to "(cell) range"? If it isn't, where such expression is defined for this context? Although I am intrigued by the term, I am not necessarily asking for an answer to be written here (welcome anyway), but I'm asking for consistency and clarity in the help files. If the term "cell range" is equivalent in this case / context (which I do _not_ know), perhaps it would be better to use a commonly-used term rather than some ambivalent less-used one, such as "domain"? * At: https://help.libreoffice.org/latest/en-US/text/scalc/01/04060104.html?DbPAR=CALC#bm_id31536851 that is: text/scalc/01/04060104.xhp Quote1: Returns the value if the cell does not contains Quote2: Returns the value if the cell does not contains In both cases (IFERROR and IFNA): "Returns the value if the cell does not contains..." ...both should be: s/contains/contain (i.e. no final "s"). -- You are receiving this mail because: You are the assignee for the bug.
[Libreoffice-bugs] [Bug 153574] RFE: Add information about errors, error codes or error numbers in Calc (that is currently not included)
https://bugs.documentfoundation.org/show_bug.cgi?id=153574 Olivier Hallot changed: What|Removed |Added Ever confirmed|0 |1 Status|UNCONFIRMED |NEW Keywords||easyHack Blocks||80430 --- Comment #1 from Olivier Hallot --- Page to fix: text/scalc/05/0214.xhp 1) Add to each entry (error code) of the table, e.g. Calc error 501 Calc error ### ... 2) add entry for N/A in table. "The formula can’t find what it’s been asked to look for." Referenced Bugs: https://bugs.documentfoundation.org/show_bug.cgi?id=80430 [Bug 80430] [META] Documentation gap for new features -- You are receiving this mail because: You are the assignee for the bug.
