pothiers / vvote Goto Github PK

Just tried: genmap -v MockSOVC.xlsx FinalLVR.xlsx mockraces mockchoices
The program appears to run but output files are empty. Input files look OK to me.

To aid feature development and adequate testing, it would help to have a small test case that contains all the variety we need to handle. This would be small like the simple files, but full of variety like the day-N* files.

The SOVC “equivalent” should include a record for each precinct taken from the LVR as well as the totals. The checking of the LVR vs SOVC could check each of these figures, not just the totals, if it doesn’t take too long.

A couple of mods needed: A real SOVC file does not have the <OOD> number in it. Also, the undervote figures should be converted to a single undervote count: undervote = undervote-1 + 2undervote-2 + 3undervote-3 + … + N*undervote-N in a vote-for-N contest. Similarly, the overvote count in an SOVC is the number of overvote entries, not the number of overvoted ballots, so the overvote figure should be multiplied by N. These differences will have to taken into account when comparing LVR to SOVC.

@ryantwaz, The real SOVC has zero counts (e.g. tests/data/mock1-sovc.xlsx). But when all I have is the CVR, there is no mention of the choices unless there are votes for them. Is there another way of getting all choices for each race?

On output report, change “” to “Ballots Cast” (#ballots - ). Would like to see “Ballots Cast” as first entry in each contest list. Undervote and overvote counts at bottom of each list.

Number of undervotes: Tom would like to keep track of the number of undervote entries for each contest. In a vote-for-N contest there could be 0,…,N “undervote” entries in each record. So define undervote-m, m=1,…,N as the number of records with m “undervote” entries.