language for description of report - and executing it (python-based). inputs, filters, aggregations, layout, output to text, pdf, html, excel, wx/preview, ..
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
engine
test
view
.gitmodules
README.md
__init__.py
aggr_func.py
common.py
extern_calc.py
field.py
field_container.py
group.py
link.py
makefile
packet.py
row.py
row_classifier.py
spravka.py

README.md

reporter

language for description of reports- and executing (python-based). inputs, filters, aggregations, layout, output to text, pdf, html, excel, wx/preview, ..


here's the situation.

  • the thing was in gonna-be-production. BUT it does not mean it's 100% stable, well-tested, documented, ... But the parent project was stopped

  • it's a shared effort by me and a colleague, me being idea-generator and big-picture/language architect, he being the implementor and everything as of lately.

  • idea of opensourcig is hanging since the beginning ~2006 (we've put about 2+ man-years into it...)

  • it needs some effort to get well-separated from the all-other-stuff, mainly configs and facading interfaces

so... it is as it is; after i describe now in brief what it is/can/might be, u decide whether u want to take the plunge and do the separation/fine-cuts - finish it, make it usable for anyone, eventualy backward compatible or not.

so, the spec in brief:

  • a declarative language for describing a report:

    • input params (this is decl. only)

    • data extraction, depending on above + extras (this is DIY)

    • data processing - filtering, calcs, aggregation/grouping/ordering - this is the main thing, the report-model

    • output layout into various looks (like MVC) - console/plaintext, wxwin/ printpreview, pdf/reportlab, html, excel, (not yet: xml, m$word, others...)

  • the modelling is discrete, that is the atomary item is a field. then fields can be grouped freeform, rows, columns, whatever

  • there is inheritance in several aspects - structure, fields, processing, layout

  • there is containment (any field can be a sub-report)

  • various report-metamodels, depending on structural flexibility

  • bunch-of-fields - free form
  • rows-of-fields - list of people
  • table - rows of (same) columns (tabular data)
  • cross-cut cols/rows (variable number rows * variable number columns)

example report:

class SprPeople( SprDef):
    class MyRow( Row):
        _set_ = FieldContainer( 
            name    = FldDef( type=Text, model='name.name',
                        view=FldView( label='Name', align='l', border='r')),
            age     = FldDef( type=Int, model='ageyears',
                        view=FldView( label='Възраст', align='r', border='r')),
            city    = FldDef( type=Text, model='address.city'),
            suburb = FldDef( type=Text, model='address.suburb'),
            )
        _calc_ = FieldContainer(
            address = FldDef( type=Text,
                        data=lambda r: r.city +' '+ r.suburb),
                        view=FldView( label='Addresse', align='l', border='r')),
                        )
        RowType = MyRow

2009, svilen dobrev & hristo stanev