Solution Finder: Path

• Guide
• Solution finder

Solution Finder’s Path command outputs all the ways to get a perfect clear from a specified field, given a specified pattern. The terminal output is written into a file in the specified log path, and a file containing the path data is generated in the specified output base.

1

java -jar sfinder.jar path --tetfu <fumen> --patterns <pattern>

1

java -jar sfinder.jar path -t <fumen> -p <pattern>

Specified Field (--tetfu, -t): the fumen code that sfinder begins working with. If not specified, the file field.txt in the input folder is used.

• Page (--page, -P): Specify the page of the fumen.
  • The default is 1 (the first page)
  • --page 2
• Clear Line (--clear-line, -c): Specify the number of line clears for a perfect clear.
  • The default is 4
  • --clear-line 6

Patterns (--patterns, -p): Determines the queues checked by sfinder. Read more about this parameter here.

• Hold (--hold, -H): Specify whether or not a hold slot is usable.
  • By default, it is use.
  • -H use or -H avoid
• Drop (--drop, -d): Specify what movements sfinder uses to try and find solutions.

Specified only (--specified-only, -so):

WIP

This section isn’t filled out yet.

Reserved minos (--reserved, -r):

WIP

This section isn’t filled out yet.

Kick table (--kicks, -K):

WIP

This section isn’t filled out yet.

Format (--format, -f): Dictates the way the path output is written onto a file.

• By default, the output is in html.
• With --format html (the default), there are two outputs (if --max-layer is not specified):
  • path_unique.html contains a list of all the possible solves found by sfinder.
  • path_minimal.html contains a loosely defined set of minimals. Read more about what these mean over at this page.
• --format csv will output the path results as a csv. You will need to specify further what kind of info will be displayed in the csv, or you will end up with nonsensical text. Some more info about different csv outputs here.

Max Layer (--max-layer, -L): refers to the outputs of path when using the html format.

• By default, it is set to 2 (outputs both path_unique.html and path_minimal.html)
• The only other option is -L 1, where it only outputs path_unique.html.

Key (--key, -k): refers to the way the path result is sorted when using the csv format.

• By default, it is set to none.
• --key solution outputs the path info grouped by solution.
• --key pattern outputs the path info grouped by queue.
• --key use outputs the path info grouped by pieces used
• The parameters may also be shortened to just the first name (--key use = -k u).

Split (--split, -s): refers to the way the solution fumen is built. - By default, it is set to no (output is a normal fumen). - --split yes builds the solves mino-by-mino (glued fumen).

Output Base (--output-base, -o): Specify the path data file output.

• By default, the path data is written into output/path.csv, path_minimal.html, or path_unique.html (depending on your path command).
• --output-base output/tubpath.csv

Log path (--log-path, -lp): Specify the .txt file output from the output of the command.

• By default, the log path is output/last_output.txt.
• --log-path output/foundpaths.txt

Specified Field from a file (--field-path, -fp): instead of defining the fumen code using --tetfu, you can specify a .txt file that contains a fumen code instead.

• By default, the field path is input/field.txt.
• --field-path input/sdpc.txt

Patterns from a file (--patterns-path, -pp): instead of defining patterns using --patterns, you can specify a .txt file that contains either the actual queues, or patterns.

• By default, the patterns path is input/patterns.txt.
• --patterns-path input/filteredqueue.txt

Threads (--threads, -th): Specify the number of threads to use when sfinder is running. Basically, threads are tiny virtual CPUs used to calculate stuff. Almost no need to ever touch this.

• By default, the number of threads used is 1.
• --threads 0 will allow sfinder to use as many threads as is present in the execution environment.

Cached Bit (--cached-bit, -cb): Specify the smallest bit of the cache to use for the internal algorithm. You most likely will never need to use this parameter.

An example output of the path command:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41

#Command Line Input
java -jar sfinder.jar path -t v115@9gD8DeF8CeG8BeH8CeC8JeAgH -p *p4

# Setup Field
XXXX____XX
XXXX___XXX
XXXX__XXXX
XXXX___XXX

# Initialize / User-defined
Max clear lines: 4
Using hold: use
Drop: softdrop
Searching patterns:
  *p4

# Initialize / System
Version = 1.0
Threads = 4
Need Pieces = 3

# Enumerate pieces
Piece pop count = 4

# Cache
  -> Stopwatch start
     ... done
  -> Stopwatch stop : avg.time = 20 ms [1 counts]

# Search
  -> Stopwatch start
     ... searching
     ... done
  -> Stopwatch stop : avg.time = 353 ms [1 counts]

# Output file
Found path [unique] = 18
Found path [minimal] = 16

# Finalize
done

These are the outputs produced by various :

1. With –format html (the default), the output looks like this for both path_minimal.html and path_unique.html:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18

18 solutions [840 input sequences]
All solutions #link

No line erasure # line erasure = line clear
J-Spawn T-Reverse I-Spawn / 11.4 % [96] # these are links to fumens
T-Spawn L-Reverse I-Spawn / 7.6 % [64]
L-Right I-Left T-Reverse / 5.7 % [48]
# 2 more lines...

With line erasure
O-Spawn J-Reverse I-Spawn / 7.6 % [64]
Z-Spawn L-Reverse I-Spawn / 7.6 % [64]
S-Spawn J-Spawn I-Spawn / 7.6 % [64]
# 10 more lines...

# "J-Spawn T-Re..." refers to mino operations based on fumen.
# "/ 11.4 %" is the percentage of queues the solution works for. (cov%)
# "[96]" is the number of queues the solution works for. (cov#)

No line erasure vs with line erasure:

2. With –format csv –key solution, the output file (output/path.csv) contains rows that have the path info sorted by solution:

1
2
3
4
5
6
7
8

v115@9gD8g0ywF8i0G8RpH8RpwwC8JeAgWDAv/1BA, #the solution
TJO, #pieces used in the solution
1, #
1, #
32, # number of queues this solution works for
OJT, #
OJT, #
SOJT;ZOJT;JOTI;OTJL;JOTS;OJTL;... # queues this solution works for

3. With –format csv –key pattern, the output file (output/path.csv) contains rows that have the path info sorted by queue:

1
2
3
4
5

OITL, # queue being checked
2, # number of solutions
TIL;TIO, # pieces used in the solutions
O;L, # pieces not used in the solutions (saved pieces)
v115@...;v115@...;... # solutions for this queue

4. With –format csv –key use, the output file (output/path.csv) contains rows that have the path info sorted by pieces used:

1
2
3
4
5

ILZ, # pieces used
1, # number of solutions that use these pieces
64, # number of queues solved with these pieces
v115@9gD8zhF8ilG8BtH8glBtC8JeAgWDA6SdBA, # solutions that use these pieces
ZSIL;TZIL;... # queues the solutions work for

1. One of path’s most important uses is as an input file for finding various types of minimals.