Thanks Mat,

ad 1
Escaping the regexp might be a solution. But how? I mean in a regular 
expression [a-z] is different from \[a-z\].
Unsure how to escape the regular expression. It might be a greediness / 
look-ahead problem.

ad 2
No luck here, tried it. If I remember correctly <<__x__>> sort of casts the 
variable into a tiddly. That might be part of the explanation.

ad 3
Sorry, typo in this mail.

Cheers,
Hylke






On Tuesday, March 2, 2021 at 11:58:52 PM UTC+1 Mat wrote:

> WOW! To say the least. (And incidentally I'm working on a distantly 
> related thing).
>
> Regarding your problems here are some first guesses:
>
> >regexp="([A-Z]+(?=[0-9]+))|([0-9]+(?=[A-Z]+))" 
> >and filter="[<ref>splitregexp<regexp>reverse[]]"
>
> Maybe you need to escape the square brackets?
>
> ><<mymacro one/2/three>> fails. I
>
> See if enclosing the argument with "..." helps
>
> ><$set filter:"filter" variable="var"> and <$vars var={{{ filter }}}> give 
> dif
>
> Again some syntactic tidbits (but maybe you just mistyped in this post):
>
> <$set filter*=*"filter" name="var">
>
>
> <:-)
>
>
>
> On Tuesday, March 2, 2021 at 10:18:33 PM UTC+1 hwva...@gmail.com wrote:
>
>> Good day,
>>
>> Triggered by the announcement of Tamasha I decided to dig into TW5 some 
>> more. Therefore I created a --javascript free -- plugin for easy import and 
>> display of spreadsheet data, or more generally dataframes. Along the way I 
>> ran into some subtleties / miscomprehensions of the tiddlywiki framework , 
>> which might be of interest to others. A demonstration of the dataframe 
>> plugin and  a working description of three subtleties in action is 
>> available at: https://hwvandijk.bitbucket.io/tw-dataframe/
>>
>> 1. splitregexp that crashes the Javascript engine of Firefox 58, but 
>> works on tiddlydesktop 0.14 (chromedriver 81)
>> I used a (elaborate) regular expression in splitregexp to split "ABC123" 
>> in to "123 ABC" in one go. 
>> i.e.
>> regexp="([A-Z]+(?=[0-9]+))|([0-9]+(?=[A-Z]+))"  
>> and filter="[<ref>splitregexp<regexp>reverse[]]"
>>
>> I get a Javascript error: *uncaught exception: Linked List only accepts 
>> string values, not undefined *
>> The problem has been solved by using a more straightforward regular 
>> expression in a  search-replace:regexp filter operator.
>>
>> 2. $x$ and <<__x__>> or similar but not identical
>> I had problems using macro parameters in filters. Therefor I cast them in 
>> a <$set> or <$vars> variable, either through $x$ or <<__x__>>. The former 
>> works fine unless a parameter x that is passed to the macro contains 
>> slashes.
>> <<mymacro one.2.three> works fine, <<mymacro one/2/three>> fails. It 
>> renders the <$vars expression> literally in the page.
>>
>> 3. <$set filter:"filter" variable="var"> and <$vars var={{{ filter }}}> 
>> give different results
>> To specify the index/columns of the dataframe  to be displayed I use a 
>> list of ranges. Meaning that a spec of "1,3-5,7" should be transformed into 
>> "1,1 3,5 7,7" such that when you feed these entries one by one into the 
>> range operator you get [1 2 3 4 5 7]. 
>> However with:
>>
>> <$vars p=<<__param__>> 
>>       regexp="^(\d+)$" > 
>> <$set name="setref" 
>> filter="[<p>split[,]search-replace:g:regexp<regexp>,[$1,$1]]" >
>> <$vars varsref={{{ [<p>split[,]search-replace:g:regexp<regexp>,[$1,$1]] 
>> }}} >
>>
>> The variables *setref* and *varsref* are not identical. *varsref* is 
>> wrong, you should use *setref*.
>> varsref only works for simple specifications, such as "7". It looks like 
>> *varsref 
>> *does not obey the g (global) specifier.
>>  
>> Sorry for the long mail, but hopefully someone can point out my 
>> misconceptions or file a bug if that is appropriate.
>>
>> Cheers,
>>
>> Hylke van Dijk
>>
>>
>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to tiddlywiki+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tiddlywiki/70c6057f-0afa-4a30-b4ce-9aa260d5a0f4n%40googlegroups.com.

Reply via email to