Delete text in a txt file

[myspacee]

Hello to all,

scripting about pdf conversion to txt. (ebook, epub, flash related)

I've some pdf, extract text, clean a bit and use for other task.

Search help to remove in txt file some text. Post some rows:

#<font 10 ""> 3d hdtv: ready for primetime? #<font 7 ""> Tablets 2.0 Why 2010 Could (Finally) Be Their Year nexus oneHan
#<font 16 ""> www.storemags.com & www.fantamag.com #<font 8 ""> MARCH 2010 MARCH 2010 vol. 29 no. 3 44CovER SToRY Though
#<font 47 ""> www.storemags.com & www.fantamag.com #<font 26 "">#<font 24 ""> PC Magazine Digital Edition, #<font 8 ""> 
#<font 24 ""> www.storemags.com & www.fantamag.com #<font 21 ""> The iPad: A Must-Have? #<font 17 ""> The New York Times

I've 2 task to accomplish:

- 1st remove text into ""> R4Nd0M #<font marker

- 2nd remove text into #<font R4Nd0M ""> marker

See that markers can contain random text (can be number or letter),

and can't find solution or function that can help me.

Any hint is appreciated, thank you.

m.

[myspacee]

up,

try to understand how use StringRegExpReplace but have some problem...

anyone can help a bit ?

Thank you,

m.

[Melba23]

myspacee,

I am confused about what you want to do - can you show us how you want the lines to look after the 2 deletions?

If you want to get rid of the 2 font tags, then this pattern should work:

StringRegExpReplace($sText, "(?U)(#<.*>)", "")

(?U) = Inverse greediness, look for the shortest match (otherwise you lose the text between the tags as well!)

(#<.*>) = Look for #<, followed by a number of characters, followed by >

M23

[myspacee]

Thank you for reply,

I've

#<font 10 ""> 3d hdtv: ready for primetime? #<font 7 ""> Tablets 2.0 Why 2010 Could (Finally) Be Their Year nexus oneHan
#<font 16 ""> www.storemags.com & www.fantamag.com #<font 8 ""> MARCH 2010 MARCH 2010 vol. 29 no. 3 44CovER SToRY Though
#<font 47 ""> www.storemags.com & www.fantamag.com #<font 26 "">#<font 24 ""> PC Magazine Digital Edition, #<font 8 ""> 
#<font 24 ""> www.storemags.com & www.fantamag.com #<font 21 ""> The iPad: A Must-Have? #<font 17 ""> The New York Times

I want to obtain

Tablets 2.0 Why 2010 Could (Finally) Be Their Year nexus oneHan
MARCH 2010 MARCH 2010 vol. 29 no. 3 44CovER SToRY Though
PC Magazine Digital Edition,
The iPad: A Must-Have? The New York Times

try few StringRegExpReplace combination but i'm not so smart as i think :]

m.

[Melba23]

myspacee,

I can do it in 2 passes.

First pass - get rid of the initial #<font tag> text #<font tag>:

StringRegExpReplace($sText, "(?U)(?m:^)(#<.+>.+#<.+>)", "")

(?U) = Inverse greediness, look for shortest match

(?m:^) = Start at beginning of a line

(#<.+>.+#<.+>) = Match the first 2 tags and any text between them

Then a second pass to get rid of the remaining #<font tag>:

StringRegExpReplace($sText, "(?U)(#<.+>)", "")

(?U) = Inverse greediness

(#<.+>) = match the remaining tags

I am sure a SRE guru will come along in a minute and laugh until his sides hurt - but that should get you started!

M23

[myspacee]

I think to use 2 step too,

but not discard #<font tag> text #<font tag> :

$chars = StringRegExpReplace($chars, "(?U)(> .*#<)", "")
$chars = StringRegExpReplace($chars, "(?U)(#<.*>)", "")

Is a bad idea ? [EDIT: after test is a very bad idea !]

m.

(now testing yours...)

Edited February 28, 2010 by myspacee