if self.id_counter > 1:
blames = self.__get_blames (read_cursor, repoid)
job_pool = JobPool (repo, path or repo.get_uri (), queuesize=100)
# Get code files
query = "select f.id from file_types ft, files f " + \
"where f.id = ft.file_id and " + \
"ft.type in ('code', 'unknown') and " + \
"f.repository_id = ?"
read_cursor.execute (statement (query, db.place_holder), (repoid,))
code_files = [item[0] for item in read_cursor.fetchall ()]
n_blames = 0
fr = FileRevs (db, cnn, read_cursor, repoid)
for revision, commit_id, file_id, action_type, composed in fr:
if file_id not in code_files:
continue
if (file_id, commit_id) in blames:
printdbg ("%d@%d is already in the database, skip it", (file_id, commit_id))
continue
if composed:
rev = revision.split ("|")[0]
else:
rev = revision
relative_path = fr.get_path ()
printdbg ("Path for %d at %s -> %s", (file_id, rev, relative_path))
# after download. However, ignore them for now to speed things up
query = "select f.id from file_types ft, files f " + \
"where f.id = ft.file_id and " + \
"ft.type in('code') and " + \
"f.repository_id = ?"
# "ft.type in('code', 'unknown') and " + \
read_cursor.execute(statement(query, db.place_holder), (repo_id, ))
code_files = [item[0] for item in read_cursor.fetchall()]
query = """select c.file_id, c.commit_id from content c, files f
where c.file_id=f.id and f.repository_id=?
"""
read_cursor.execute(statement(query, db.place_holder), (repo_id, ))
existing_content = [(item[0], item[1]) \
for item in read_cursor.fetchall()]
fr = FileRevs(db, connection, read_cursor, repo_id)
i = 0
# Loop through each file and its revision
for revision, commit_id, file_id, action_type, composed in fr:
# loop_start = datetime.now()
if file_id not in code_files:
continue
if (file_id, commit_id) in existing_content:
continue
try:
relative_path = fr.get_path()
except AttributeError, e:
printerr("No path found for file %d at commit %d",
(file_id, commit_id))
# after download. However, ignore them for now to speed things up
query = "select f.id from file_types ft, files f " + \
"where f.id = ft.file_id and " + \
"ft.type in('code') and " + \
"f.repository_id = ?"
# "ft.type in('code', 'unknown') and " + \
read_cursor.execute(statement(query, db.place_holder), (repo_id,))
code_files = [item[0] for item in read_cursor.fetchall()]
query = """select c.file_id, c.commit_id from content c, files f
where c.file_id=f.id and f.repository_id=?
"""
read_cursor.execute(statement(query, db.place_holder), (repo_id,))
existing_content = [(item[0], item[1])
for item in read_cursor.fetchall()]
fr = FileRevs(db, connection, read_cursor, repo_id)
i = 0
# Loop through each file and its revision
for revision, commit_id, file_id, action_type, composed in fr:
if action_type == 'D':
continue
# loop_start = datetime.now()
if file_id not in code_files:
continue
if (file_id, commit_id) in existing_content:
continue
try:
relative_path = fr.get_path(repo, path or repo.get_uri())
except TypeError as e:
query = (
"select f.id from file_types ft, files f "
+ "where f.id = ft.file_id and "
+ "ft.type in('code') and "
+ "f.repository_id = ?"
)
# "ft.type in('code', 'unknown') and " + \
read_cursor.execute(statement(query, db.place_holder), (repo_id,))
code_files = [item[0] for item in read_cursor.fetchall()]
query = """select c.file_id, c.commit_id from content c, files f
where c.file_id=f.id and f.repository_id=?
"""
read_cursor.execute(statement(query, db.place_holder), (repo_id,))
existing_content = [(item[0], item[1]) for item in read_cursor.fetchall()]
fr = FileRevs(db, connection, read_cursor, repo_id)
i = 0
# Loop through each file and its revision
for revision, commit_id, file_id, action_type, composed in fr:
if action_type == "D":
continue
# loop_start = datetime.now()
if file_id not in code_files:
continue
if (file_id, commit_id) in existing_content:
continue
try:
relative_path = fr.get_path(repo, path or repo.get_uri())
except TypeError as e:
if self.id_counter > 1:
blames = self.__get_blames(read_cursor, repoid)
job_pool = JobPool(repo, path or repo.get_uri(), queuesize=100)
# Get code files
query = "select f.id from file_types ft, files f " + \
"where f.id = ft.file_id and " + \
"ft.type in ('code', 'unknown') and " + \
"f.repository_id = ?"
read_cursor.execute(statement(query, db.place_holder), (repoid, ))
code_files = [item[0] for item in read_cursor.fetchall()]
n_blames = 0
fr = FileRevs(db, cnn, read_cursor, repoid)
for revision, commit_id, file_id, action_type, composed in fr:
if file_id not in code_files:
continue
if (file_id, commit_id) in blames:
printdbg("%d@%d is already in the database, skip it",
(file_id, commit_id))
continue
if composed:
rev = revision.split("|")[0]
else:
rev = revision
relative_path = fr.get_path(repo, path or repo.get_uri())
