Instruct no uuid module

Author: colinmcneil

prompts/pylint/2-make-db-schema.md

@@ -35,12 +35,10 @@ tools:
 delete the /thread/db.sqlite 
 
 Write a SQL schema for a table named RANGES with columns ID, PATH, START_LINE, END_LINE, START_COLUMN, and 
-END_COLUMN. The ID is the primary key and auto increments.
+END_COLUMN. The ID is going to be a unique string.
 
-Write a SQL schema for a table named VIOLATIONS with a foreign key named RANGE of type ID, STRING, and columns
-MESSAGE, TYPE, VIOLATION_ID. The ID is the primary key, and auto increments.
-
-allow END_LINE and END_COLUMN to be nullable. 
+Write a SQL schema for a table named VIOLATIONS which references RANGES with fk `RANGE_ID`, and columns
+ID, MESSAGE, TYPE, and VIOLATION_ID. The ID is the primary key, and auto increments.
 
 Run the sqlite command with the database set to `/thread/db.sqlite` and send the SQL generated above.
 

prompts/pylint/3-generate-violation-insert.md

@@ -22,16 +22,19 @@ It should then iterate over each element of an array with the following schema:
 ```json
 [
   {"message": "some violation", "path": "app.py", "type": "error",
-   "line": 0, "column": 0, "endLine": 1, "endColumn": 1, "path": "src/app.py"}
+   "line": 0, "column": 0, "endLine": None, "endColumn": 1, "path": "src/app.py"}
 ]
 
 ```
 
 For each element of the array, it should create two INSERT statements. 
-The first should insert the columns PATH, START_LINE, END_LINE, START_COLUMN, END_COLUMN 
-into a tabled named RANGES using the properties from the map.
-The second should insert the columns MESSAGE, TYPE, RANGE, VIOLATION_ID into a table named VIOLATIONS 
-using the properties from the map.  The RANGE column should be the ID of the previous row in the RANGES table.
+
+1. The first should insert the columns ID,PATH, START_LINE, END_LINE, START_COLUMN, END_COLUMN 
+into a tabled named RANGES using the properties from the map. ID should be a random string. No uuid module here.
+2. The second should insert the columns MESSAGE, TYPE, RANGE_ID, VIOLATION_ID into a table named VIOLATIONS using the properties from the map. The RANGE_ID will be the same random string from before. 
+
+If the violation doesn't have an `endLine` or `endColumn` you should simply re-use the start position.
+
 If any strings contain single quotes, they should be escaped.
 
 Line and column represent START_LINE and START_COLUMN

prompts/pylint/4-run-violation-insert.md

@@ -49,7 +49,7 @@ docker run -it --rm -v thread:/thread \
 docker run -it --rm -v thread:/thread \
            vonwig/sqlite:latest \
            /thread/db.sqlite \
-           "SELECT * FROM VIOLATIONS INNER JOIN RANGES ON VIOLATIONS.RANGE = RANGES.ID WHERE RANGES.PATH = 'src/app.py'"
+           "SELECT * FROM VIOLATIONS INNER JOIN RANGES ON VIOLATIONS.RANGE_ID = RANGES.ID WHERE RANGES.PATH = 'src/app.py'"
 ```
 
 ```

prompts/pylint/5-1-insert-top-level-ranges.md

@@ -118,14 +118,16 @@ An example parsed file:
   ``` 
   to get the ranges of each top level node into capturing groups.
   Write the ranges to `/thread/ranges.sql` in the form of sql 
-  `INSERT INTO RANGES (START_LINE, END_LINE, START_COLUMN, END_COLUMN, PATH) values (...)`
+  `INSERT INTO RANGES (ID, START_LINE, END_LINE, START_COLUMN, END_COLUMN, PATH) values (...)`
+
+You need to generate a random string without importing any modules.
 
 Example output of script would be the creation of file `/thread/ranges.sql` with contents:
 
 ```sql
-INSERT INTO RANGES (START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values (0,0,0,21, 'main.py')
-INSERT INTO RANGES (START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values (2,0,2,37, 'main.py')
-INSERT INTO RANGES (START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values (4,0,5,17, 'main.py')
+INSERT INTO RANGES (ID, START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values ('<random-string>', 0,0,0,21, 'main.py')
+INSERT INTO RANGES (ID, START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values ('<random-string>', 2,0,2,37, 'main.py')
+INSERT INTO RANGES (ID, START_LINE, START_COLUMN, END_LINE, END_COLUMN, PATH) values ('<random-string>', 4,0,5,17, 'main.py')
 ```
 
 5. Execute the code in the javascript sandbox.

prompts/pylint/5-2-insert-code.md

@@ -61,7 +61,7 @@ where query selects everything from the RANGES table where END_LINE is not null.
 we get something like
 ```csv
 ID,PATH,START_LINE,END_LINE,START_COLUMN,END_COLUMN
-125,wsgi_skylines.py,11,104,0,61
+1234-abcd-1234,wsgi_skylines.py,11,104,0,61
 ```
 
 So, write a JS script to 
@@ -75,6 +75,6 @@ The CODE_AT_RANGE is the sliced content at PATH
 Example `/thread/code_insert/insert_code.sql` file:
 
 ```sql
-INSERT INTO CODE (RANGE_ID, CODE_AT_RANGE) (125, "<Contents of wsgi_skylines from line 11 to 104>")
+INSERT INTO CODE (RANGE_ID, CODE_AT_RANGE) ('1234-abcd-1234', "<Contents of wsgi_skylines from line 11 to 104>")
 ```
 

prompts/pylint/generic_query.md

@@ -20,4 +20,19 @@ tools:
 
 # prompt user
 
-How many rows are in the ranges and violations tables? Use database `/thread/db.sqlite`
+The database is `/thread/db.sqlite`
+
+Tables:
+CODE
+RANGES
+VIOLATIONS
+
+Both violations and code tables share a foreign key reference to RANGES.
+
+Query the shcema of all of these tables.
+
+Answer the following questions:
+
+What is the last violation with a non-null END_LINE in ranges?
+
+What is the code for that violation?