Repo layout

[davidlinsley]

I dumped all 341 scanned pages into the OriginalScans directory. Given we have commit history and moves are just cheap renames, I'm suggesting the following new layout as the next commit:

• OriginalScans
  • <filename.sa>
    • Scanned files for filename.sa

For example:

• OriginalScans
  • ECBRAM.SA
    • CCI10222022_0001.PNG
    • ...
    • CCI10222022_0008.PNG
  • ECBBAS.SA
    • CCI10222022_0009.PNG
    • ...
    • CCI10222022_0056.PNG
  • ...

Once that is committed we create a similar OCR directory structure to dump the OCR of each page:

• OCR
  • ECBRAM.SA
    • CCI102220_0001.TXT
    • ...
    • CCI102220_0008.TXT
  • ECBBAS.SA
    • CCI10222022_0009.TXT
    • ...
    • CCI10222022_0056.TXT
  • ...

As source files are reconstructed from the OCR, they are laid out like:

• src
  • Core
    • ECBCOM.SA
    • ECBRAM.SA
    • ECBBAS.SA
    • ECBBS2.SA
    • ECBTAB.SA
    • ECBDNL.SA
    • ECBPU.SA
    • ECBMTH.SA
    • ECBEXT.SA
    • ECBGRP.SA
  • Customized
    • ECBMOD.SA
    • ECBM64.SA
  • Dragon Data
    • ECBOEM.SA
    • ECBO64.SA

Note that ECBCOM.SA will be created from the first files lines of the first page of each other .SA file.

[davidlinsley]

I should write a note about the layout of the listing, which does include the assembled bytes. We should consider reconstructing the files in two forms, one with all the contents of each line (just remove the title line) and then later cropping the first X characters of each line so it is just the LABEL OPCODE COMMENT.