On Tue, Feb 4, 2020 at 11:52 AM Cyrille <lafricai...@gmail.com> wrote:
> Hello Greg, > Can you give more information about this python library please. It's > interesting. How to use it? > The Python library is a binding of the C++ library that is auto-generated with Swig. So its API is almost the exact same as the C++ library, with a tiny number of additional bits to smooth the way into the Python world. In general, if it happens in the C++ code, you can rely on the same classes, objects, and methods to exist in the Python bindings. Even most of the operator definitions are maintained, although not all of them are possible as you are more limited in how you express those in Python. As I'm not an expert on the C++ API, any particular details you will need to ask those more knowledgeable about. But you should be able to scan any C or C++ Sword code and directly translate the calls into Python. Is the library in the linux repo? > That is going to be distro dependent. I maintain it in Fedora 31 as "python3-sword" (and previous as python2-sword and python-sword before that). I believe it's also in the EPEL7 repository for CentOS/RHEL 7 users. It might be in EPEL8, if that's your thing, as well, but if not let me know and I'll make the branch for that. Other distros, you'll have to check. As long as your distro includes Python 2 or 3 build headers and the Swig tool (most of them do), then building it shouldn't be difficult. My build tree for Fedora is here: https://src.fedoraproject.org/rpms/sword/tree/master. To build the same either use SVN HEAD, or use my two swig-related patches in that tree, and add the appropriate options to your CMake invocation (they can be found in the sword.spec file but amount to -DSWORD_PYTHON_3:BOOL=TRUE to build the Python 3 version). --Greg > Le 04/02/2020 à 12:41, Greg Hellings a écrit : > > Maxwell, > > If you install the Python bindings to the Sword library, you can use the > library's extensive parsing information as well as its knowledge of > locales. A very simple Python script[0] will iterate all lines of input > (you can give it a list of file arguments, you can pipe the output of a > different program to it, you can write the lines in manually from stdin) > and parse them. Doing exactly this work was impetus to get the bindings > fixed up and compiling again some years back when converting references by > external means was awfully slow for another member of this list. Using the > bindings like this became nearly fool-proof and brought down the amount of > time required to execute from unbearably long periods to under a second. > > --Greg > > [0] https://gist.github.com/greg-hellings/0de55fc3e07d5014f005efc12ffbdffa > > On Tue, Feb 4, 2020 at 11:28 AM Maxwell Murunga <maxm...@gmail.com> wrote: > >> Thank you Dominique; Thanks Cyrille; Thanks Greg. >> >> >> >> # “Additional” steps to get the awk script >> >> # working fine on macOS as it does on Linux >> >> $ brew install gawk >> >> >> >> # If Terminal Throws Error >> >> $ brew unlink awk >> >> $ brew link --overwrite gawk >> >> >> >> # Confirm all went well! >> >> $ gawk --version >> >> >> >> # Now proceed as normal >> >> # Make the executable >> >> $ chmod +x Ref2Osis.sh >> >> >> >> # Thereafter, run it >> >> $ ./Ref2Osis.sh >> >> >> Works Perfect. >> >> >> Blessed [be] the LORD God of Israel from everlasting, and to everlasting. >> Amen, and Amen. >> >> >> >> ~~Shalom. >> >> On Sun, Feb 2, 2020 at 3:39 AM Cyrille <lafricai...@gmail.com> wrote: >> >>> What are you doing exactly? If you try to convert the ref to osisRef >>> Dominique wrote an awk script which works pretty good. >>> See the attached file. >>> >>> Le 01/02/2020 à 18:06, Maxwell Murunga a écrit : >>> >>> Greetings Saints, >>> >>> I'm processing an OSIS Commentary in InDesign using GREP: >>> >>> *Find: *((\d+ )?(\w+?.? \d+[:]\d+)(.\d+)?([, \d]+(.\d+)?)*) >>> *Replace:* <reference osisRef="$1">$1</reference> >>> >>> It partially accomplishes the task, but does not automatically convert >>> the book names to the standard OSIS abbreviations. I also need help in >>> figuring out how to add looking for Arabic and Roman numerals (1-2 >>> instances of the letter "I"; or simply "1" or "2" ) to cover instances of >>> something like I Corinthians or II Corinthians; 1 Corinthians or 2 >>> Corinthians. >>> >>> Could anyone be so kind enough as to provide a *grep* or *sed* script >>> to auto convert any kind of Bible reference into this format: >>> >>> <reference osisRef="Gen.1.1">Genesis 1:1</reference> >>> <reference osisRef="2Chr.1.1">2 Chronicles 1:1</reference> >>> <reference osisRef="2Chr.1.1">II Chronicles 1:1</reference> >>> >>> In Christ Alone, >>> >>> Maxwell. >>> >>> _______________________________________________ >>> sword-devel mailing list: >>> sword-devel@crosswire.orghttp://www.crosswire.org/mailman/listinfo/sword-devel >>> Instructions to unsubscribe/change your settings at above page >>> >>> >>> _______________________________________________ >> sword-devel mailing list: sword-devel@crosswire.org >> http://www.crosswire.org/mailman/listinfo/sword-devel >> Instructions to unsubscribe/change your settings at above page > > >
_______________________________________________ sword-devel mailing list: sword-devel@crosswire.org http://www.crosswire.org/mailman/listinfo/sword-devel Instructions to unsubscribe/change your settings at above page