# More search and replace

**URL:** <https://www.ruby-forum.com/t/more-search-and-replace/78008>\
**Category:** Ruby\
**Created:** [December 2, 2006, 7:51pm UTC](https://www.ruby-forum.com/t/more-search-and-replace/78008 "2006-12-02T19:51:58Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![ishamid](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/ad7895/32.png) [@ishamid](https://www.ruby-forum.com/u/ishamid)\
**Post date:** [December 2, 2006, 7:51pm UTC](https://www.ruby-forum.com/t/more-search-and-replace/78008/1 "2006-12-02T19:51:58Z")

</div>

[Total novice]

A follow-up on my last email (“search and replace”)". I am trying to  
convert an OOo xml source (content.xml) to TeX. It’s a bibliography and  
thus very predictable/regular/simple etc. Each entry looks roughly like  
this (simplified):

====================================  
\<text:p text:style-name=“ID”\>[\<text:sequence text:ref-name=“refAutoNr3”

```
text:name="AutoNr" text:formula="ooow:AutoNr+1"

```

# style:num-format=“1”\>4\</text:sequence\>\</text:p\> \<text:p text:style-name=“Standard”\>Ben\</text:p\> \<text:p text:style-name=“reference”\> \<text:span text:style-name=“T10”\>Article\</text:span\>., \<text:span text:style-name=“Style2”\>Journal\</text:span\>, volume, issue, year. \</text:p\> \<text:p text:style-name=“reference”/\> \<text:p text:style-name=“reference”/\>

I. line one is discussed in my last email. Basically, each line of this  
type (numbers are variable) needs to be converted to

# ==== \head

# II.

# \<text:p text:style-name=“P6”\>Jim\</text:p\> \<text:p text:style-name=“P8”\>Michael\</text:p\> \<text:p text:style-name=“Standard”\>Ben\</text:p\>

replace each with the name plus a linespace

====================================  
Jim

Michael

# Ben

III. \<text:span text:style-name=“T10”\>Article\</text:span\>

If the style-name=“T10”, then the argument should be, e.g. {\bf  
Article}  
if the style-name=“Style2”, then argument should be, e.g. {\it  
Journal}

IV. So the final output should be something like

====================================  
\head Ben

{\bf Article}, {\it Journal}, volume, issue, year.

====================================

I hope to get enough info here to be able to finish this myself. I  
assume finishing my script would only take one of you guys 15 or 20  
minutes 😉 If I’m not able to get things working quickly (trying to  
learn Ruby and do my work at the same time) I will be happy to pay one  
of you for an hour or so of work (I’m up against a deadline).

THANK YOU  
Idris

PS For reference, here is the script I’m trying to modify for this OOo  
bibliography:

=====================================  
class OpenOffice

```
# using an xml parser if overkill and we need to regexp anyway

attr_reader :display, :inline, :translate
attr_writer :display, :inline, :translate

def initialize
    @data = nil
    @file = ''
    @display = Hash.new
    @inline = Hash.new
    @translate = Hash.new
end

def load(filename)
    if not filename.empty? and FileTest.file?(filename) then
        begin
            @data, @file = IO.read(filename), filename
        rescue
            @data, @file = nil, ''
        end
    else
        @data, @file = nil, ''
    end
end

def save(filename='')
    if filename.empty? then
        filename = "clean-#{@file}"
    end
    if f = open(filename,'w') then
        f.puts(@data)
        f.close
    end
end

def convert
    @translations = Hash.new
    @translate.each do |k,v|
        @translations[/#{k}/] = v
    end
    if @data then
        @data.gsub!(/<\?.*?\?>/) do
            # remove
        end
        @data.gsub!(/<!--.*?-->/) do
            # remove
        @data.gsub!(/<!--.*?-->/) do
            # remove
        end
        @data.gsub!(/.*?<(office:text).*?>(.*?)<\/\1>.*/mois) do
            '\starttext' + "\n" + $2 + "\n" + '\stoptext'
        end

```

@data.gsub!(/\<(office:font-face-decls|office:automatic-styles|text:sequence-decls)._?\>._?\</\1\>/mois)  
do  
# remove  
end

@data.gsub!(/text:span.\*?text:style-name=([\’\"])(.\*?)\1(._?)\</text:span\>/)  
do  
tag, text = $2, $3  
if inline[tag] then  
(inline[tag][0]||’’) + clean\_display(text) +  
(inline[tag][1]||’’)  
else  
clean\_display(text)  
end  
end  
 @data.gsub!(/text:p[^]_?/\>/) do  
# remove  
end

@data.gsub!(/text:p.\*?text:style-name=([\’\"])(.\*?)\1(.\*?)\</text:p\>/)  
do  
tag, text = $2, $3  
if display[tag] then  
“\n” + (display[tag][0]||’’) + clean\_inline(text) +  
(display[tag][1]||’’) + “\n”  
else  
“\n” + clean\_inline(text) + “\n”  
end  
end  
@data.gsub!(/\t/,’ ‘)  
@data.gsub!(/^ +$/,’’)  
@data.gsub!(/\n\n+/moi,"\n\n")  
end  
end

```
def clean_display(str)
    str.gsub!(/&quot;(.*?)&quot;/) do
        '\quotation {' + $1 + '}'
    end
    str
end

def clean_inline(str)
    @translations.each do |k,v|
        str.gsub!(k,v)
    end
    str
end

```

end

def convert(filename)

```
doc = OpenOffice.new

doc.display['P1'] = ['\chapter{','}']
doc.display['P2'] = ['\startparagraph'+"\n","\n"+'\stopparagraph']
doc.display['P3'] = doc.display['P2']

doc.inline['T1'] = ['','']
doc.inline['T2'] = ['{\sl ','}']

doc.translate['¬'] = 'XX'
doc.translate['&apos;'] = '`'

doc.load(filename)

doc.convert

doc.save

```

end

filename = ARGV[0]

filename = ‘content.xml’ if not filename or filename.empty?

# convert(‘content.xml’)

---

<div class="post-metadata">

**Author:** ![ishamid](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/ad7895/32.png) [@ishamid](https://www.ruby-forum.com/u/ishamid)\
**Post date:** [December 2, 2006, 7:57pm UTC](https://www.ruby-forum.com/t/more-search-and-replace/78008/2 "2006-12-02T19:57:59Z")

</div>

Are you using OOo 2.0.4? I know it has a TeX/BibTeX export feature  
now…

It’s not Ruby, but it should work (unless you’re using this with some  
sort of automated system). 🙂

–Jeremy

---

<div class="post-metadata">

**Author:** ![ishamid](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/ad7895/32.png) [@ishamid](https://www.ruby-forum.com/u/ishamid)\
**Post date:** [December 2, 2006, 8:26pm UTC](https://www.ruby-forum.com/t/more-search-and-replace/78008/3 "2006-12-02T20:26:02Z")

</div>

Hi Jeremy,

On Dec 2, 11:57 am, “Jeremy McAnally” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)  
wrote:

> Are you using OOo 2.0.4? I know it has a TeX/BibTeX export feature now…

Wow, I did not know this, but…

> It’s not Ruby, but it should work (unless you’re using this with some  
> sort of automated system). 🙂

I use ConTeXt, not LaTeX, and the two are really different, so…

I am sending a note to the ConTeXt developers list about this; maybe  
some of them can port the OOo LaTeX filters to ConTeXt. In the meantime  
I think it’s best to finish that script…

Thank you very much for letting me know about OOo and LaTeX!

Best  
Idris

---

<div class="post-metadata">

**Author:** ![ishamid](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/i/ad7895/32.png) [@ishamid](https://www.ruby-forum.com/u/ishamid)\
**Post date:** [December 2, 2006, 8:52pm UTC](https://www.ruby-forum.com/t/more-search-and-replace/78008/4 "2006-12-02T20:52:09Z")

</div>

On Dec 2, 12:21 pm, “ishamid” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> Hi Jeremy,
> 
> On Dec 2, 11:57 am, “Jeremy McAnally” [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)  
> wrote:
> 
> > Are you using OOo 2.0.4? I know it has a TeX/BibTeX export feature now…Wow, I did not know this, but…
> 
> > It’s not Ruby, but it should work (unless you’re using this with some  
> > sort of automated system). 🙂

I checked it out; the source is way too messy for my purposes; it will  
be much easier to convert the xml to ConTeXt than the LaTeX to ConTeXt.

Thnx again  
Idris