将提交详细信息的字符串列表转换为按名称和日期应用分组的结构化字典
Transform list of strings of commit-details to structured dictionary applying grouping by name and date
根据我的数据，我想以这样的形式显示，其中提交键将有一个数组
在特定日期完成的提交。这就是我期望的输出
{
"Dan Ab": [
{
"2014-05-2": {
"commit_count": "1",
"commit": [{ 'commit_hash': {'lines_added': 10, 'lines_removed': 4 }}]
},
"2014-05-3": {
"commit_count": "2",
"commit": [
{ 'commit_hash': {'lines_added': 10, 'lines_removed': 4 }},
{ 'commit_hash': {'lines_added': 14, 'lines_removed': 0 }},
]
},
}
],
"John": [
"2020-10-14": {
"commit_count": "1",
"commit": [{ 'commit_hash': {'lines_added': 1740, 'lines_removed': 10 }}]
}
]
}
但是，同一日期会多次显示，而不是像上面那样为特定日期和特定作者附加提交相关信息。
这是我的做法，但行不通
import re
merged_result = [
"43f4cc160;Dan Ab;2021-06-17; 1 file changed, 10 insertions(+), 19 deletions(-)",
"6cbf2a8b3;Dan Ab;2021-06-15; 1 file changed, 14303 insertions(+)",
"c0a77029c;Dan Ab;2021-06-15; 1 file changed, 1 insertion(+), 1 deletion(-)",
"f283d7524;Dan Ab;2021-06-15; 1 file changed, 5260 deletions(-)",
"03c5314b4;Dan Ab;2021-06-15; 5 files changed, 5265 insertions(+), 12690 deletions(-)",
"daf38ecdf;Dan Ab;2020-12-11; 1 file changed, 8 insertions(+)",
"b5eabd543;Dan Ab;2020-10-14; 1 file changed, 17 insertions(+)",
"6d50a9d09;Dan Ab;2020-10-14; 43 files changed, 15740 insertions(+), 1 deletion(-)",
"7d59n9d09;John;2020-10-14; 4 files changed, 1740 insertions(+), 10 deletion(-)"
]
coding_days = {}
total_lines = 0
total_lines_added = 0
total_lines_removed = 0
total_files_changed = 0
def getstatsummarycounts(line):
"""
1 file changed, 5 insertions(+), 1 deletion(-) - returns ['1', '5', '1']
"""
numbers = re.findall("\d+", line)
if len(numbers) == 1:
# neither insertions nor deletions: may probably only happen
# for "0 files changed"
numbers.append(0)
numbers.append(0)
elif len(numbers) == 2 and line.find("(+)") != -1:
numbers.append(0) # only insertions were printed on line
elif len(numbers) == 2 and line.find("(-)") != -1:
numbers.insert(1, 0) # only deletions were printed on line
return numbers
for result in merged_result:
[commit_hash, author, commit_date, logs] = result.split(";")
numbers = getstatsummarycounts(logs)
if len(numbers) == 3:
(files_changed, inserted, deleted) = map(lambda el: int(el), numbers)
total_lines += inserted
total_lines -= deleted
total_lines_added += inserted
total_lines_removed += deleted
total_files_changed += files_changed
if author not in coding_days:
coding_days[author] = []
else:
if commit_date not in coding_days[author]:
coding_days[author].append({commit_date: []})
else:
coding_days[author][0][commit_date].append({
commit_hash: {
"lines_added": inserted,
"lines_deleted": deleted,
}
})
else:
(files_changed, inserted, deleted) = (0, 0, 0)
我也创建了一个副本，这里是
您可以像这样解析和重组您的数据：
merged_result = [
"43f4cc160;Dan Ab;2021-06-17; 1 file changed, 10 insertions(+), 19 deletions(-)",
"6cbf2a8b3;Dan Ab;2021-06-15; 1 file changed, 14303 insertions(+)",
"c0a77029c;Dan Ab;2021-06-15; 1 file changed, 1 insertion(+), 1 deletion(-)",
"f283d7524;Dan Ab;2021-06-15; 1 file changed, 5260 deletions(-)",
"03c5314b4;Dan Ab;2021-06-15; 5 files changed, 5265 insertions(+), 12690 deletions(-)",
"daf38ecdf;Dan Ab;2020-12-11; 1 file changed, 8 insertions(+)",
"b5eabd543;Dan Ab;2020-10-14; 1 file changed, 17 insertions(+)",
"6d50a9d09;Dan Ab;2020-10-14; 43 files changed, 15740 insertions(+), 1 deletion(-)",
"6d50a9d09;Dan Ab;2020-10-14; Steak and Fries, no Salad",
]
节目：
import re
grouped = {}
pattern = r"(\d+) file[^,]*(?:\, (\d+) ins[^,]+)?(?:\, (\d+) del.+)?$"
for line in merged_result:
tag, name, date, changes = line.split(";", 3)
try:
# this will throw "NoneType" has no .groups() if not matched
files, inserts, deletes = re.search(pattern, changes).groups()
inserts, deletes = inserts or "0", deletes or "0"
except AttributeError as a:
print("Skipping: '",line,
"': cannot match data by regex to get changed/inserted/deleted\n", a)
continue
nameDict = grouped.setdefault(name, {})
dateDict = nameDict.setdefault(date, {})
dateDict.setdefault("commit_count", 0)
dateDict["commit_count"] += 1
commList = dateDict.setdefault("commit", [])
commList.append({"commit_hash": {"tag": tag, "files": files,
"lines_added": inserts, "lines_removed": deletes}})
print(grouped)
输出：
Skipping: ' 6d50a9d09;Dan Ab;2020-10-14; Steak and Fries, no Salad ': cannot match data by regex to get changed/inserted/deleted
{'Dan Ab': {'2021-06-17': {'commit_count': 1, 'commit': [{'commit_hash': {'tag': '43f4cc160', 'files': '1', 'lines_added': '10', 'lines_removed': '19'}}]}, '2021-06-15': {'commit_count': 4, 'commit': [{'commit_hash': {'tag': '6cbf2a8b3', 'files': '1', 'lines_added': '14303', 'lines_removed': '0'}}, {'commit_hash': {'tag': 'c0a77029c', 'files': '1', 'lines_added': '1', 'lines_removed': '1'}}, {'commit_hash': {'tag': 'f283d7524', 'files': '1', 'lines_added': '0', 'lines_removed': '5260'}}, {'commit_hash': {'tag': '03c5314b4', 'files': '5', 'lines_added': '5265', 'lines_removed': '12690'}}]}, '2020-12-11': {'commit_count': 1, 'commit': [{'commit_hash': {'tag': 'daf38ecdf', 'files': '1', 'lines_added': '8', 'lines_removed': '0'}}]}, '2020-10-14': {'commit_count': 2, 'commit': [{'commit_hash': {'tag': 'b5eabd543', 'files': '1', 'lines_added': '17', 'lines_removed': '0'}}, {'commit_hash': {'tag': '6d50a9d09', 'files': '43', 'lines_added': '15740', 'lines_removed': '1'}}]}}}
结果字典 reformatted（第一次点击 google）：
{
"Dan Ab":{
"2021-06-17":{
"commit_count":1,
"commit":[
{
"commit_hash":{
"tag":"43f4cc160",
"files":"1",
"lines_added":"10",
"lines_removed":"19"
}
}
]
},
"2021-06-15":{
"commit_count":4,
"commit":[
{
"commit_hash":{
"tag":"6cbf2a8b3",
"files":"1",
"lines_added":"14303",
"lines_removed":"0"
}
},
{
"commit_hash":{
"tag":"c0a77029c",
"files":"1",
"lines_added":"1",
"lines_removed":"1"
}
},
{
"commit_hash":{
"tag":"f283d7524",
"files":"1",
"lines_added":"0",
"lines_removed":"5260"
}
},
{
"commit_hash":{
"tag":"03c5314b4",
"files":"5",
"lines_added":"5265",
"lines_removed":"12690"
}
}
]
},
"2020-12-11":{
"commit_count":1,
"commit":[
{
"commit_hash":{
"tag":"daf38ecdf",
"files":"1",
"lines_added":"8",
"lines_removed":"0"
}
}
]
},
"2020-10-14":{
"commit_count":2,
"commit":[
{
"commit_hash":{
"tag":"b5eabd543",
"files":"1",
"lines_added":"17",
"lines_removed":"0"
}
},
{
"commit_hash":{
"tag":"6d50a9d09",
"files":"43",
"lines_added":"15740",
"lines_removed":"1"
}
}
]
}
}
}
r"(\d+) file[^,]*(?:\, (\d+) ins[^,]+)?(?:\, (\d+) del.+)?$" 的解释：
我正在使用 re.search 所以它不一定从字符串的开头开始 - 我正在寻找：
(\d+) file[^,] a number followed by "file" consuming
anything up to (excluding) the next ","
capturing the number in a group
(?:\, (\d+) ins[^,]+)?
(?:\, (\d+) del.+)? are similar: 0 to 1 occurence of ", "
followed by a captured number
followed by a space and some text
after "ins" we capture anyting up to excluding
the next ","
after "del" we simply capture anything
$ followed by end of string
如果不存在，可选组将导致 None，因此使用 inserts, deletes = inserts or "0", deletes or "0"
将它们转换为 0
如果您需要更快的速度，您可以使用defaultdict()，但dict.setdefault 也可以。
根据我的数据，我想以这样的形式显示，其中提交键将有一个数组 在特定日期完成的提交。这就是我期望的输出
{
"Dan Ab": [
{
"2014-05-2": {
"commit_count": "1",
"commit": [{ 'commit_hash': {'lines_added': 10, 'lines_removed': 4 }}]
},
"2014-05-3": {
"commit_count": "2",
"commit": [
{ 'commit_hash': {'lines_added': 10, 'lines_removed': 4 }},
{ 'commit_hash': {'lines_added': 14, 'lines_removed': 0 }},
]
},
}
],
"John": [
"2020-10-14": {
"commit_count": "1",
"commit": [{ 'commit_hash': {'lines_added': 1740, 'lines_removed': 10 }}]
}
]
}
但是，同一日期会多次显示，而不是像上面那样为特定日期和特定作者附加提交相关信息。
这是我的做法，但行不通
import re
merged_result = [
"43f4cc160;Dan Ab;2021-06-17; 1 file changed, 10 insertions(+), 19 deletions(-)",
"6cbf2a8b3;Dan Ab;2021-06-15; 1 file changed, 14303 insertions(+)",
"c0a77029c;Dan Ab;2021-06-15; 1 file changed, 1 insertion(+), 1 deletion(-)",
"f283d7524;Dan Ab;2021-06-15; 1 file changed, 5260 deletions(-)",
"03c5314b4;Dan Ab;2021-06-15; 5 files changed, 5265 insertions(+), 12690 deletions(-)",
"daf38ecdf;Dan Ab;2020-12-11; 1 file changed, 8 insertions(+)",
"b5eabd543;Dan Ab;2020-10-14; 1 file changed, 17 insertions(+)",
"6d50a9d09;Dan Ab;2020-10-14; 43 files changed, 15740 insertions(+), 1 deletion(-)",
"7d59n9d09;John;2020-10-14; 4 files changed, 1740 insertions(+), 10 deletion(-)"
]
coding_days = {}
total_lines = 0
total_lines_added = 0
total_lines_removed = 0
total_files_changed = 0
def getstatsummarycounts(line):
"""
1 file changed, 5 insertions(+), 1 deletion(-) - returns ['1', '5', '1']
"""
numbers = re.findall("\d+", line)
if len(numbers) == 1:
# neither insertions nor deletions: may probably only happen
# for "0 files changed"
numbers.append(0)
numbers.append(0)
elif len(numbers) == 2 and line.find("(+)") != -1:
numbers.append(0) # only insertions were printed on line
elif len(numbers) == 2 and line.find("(-)") != -1:
numbers.insert(1, 0) # only deletions were printed on line
return numbers
for result in merged_result:
[commit_hash, author, commit_date, logs] = result.split(";")
numbers = getstatsummarycounts(logs)
if len(numbers) == 3:
(files_changed, inserted, deleted) = map(lambda el: int(el), numbers)
total_lines += inserted
total_lines -= deleted
total_lines_added += inserted
total_lines_removed += deleted
total_files_changed += files_changed
if author not in coding_days:
coding_days[author] = []
else:
if commit_date not in coding_days[author]:
coding_days[author].append({commit_date: []})
else:
coding_days[author][0][commit_date].append({
commit_hash: {
"lines_added": inserted,
"lines_deleted": deleted,
}
})
else:
(files_changed, inserted, deleted) = (0, 0, 0)
我也创建了一个副本，这里是
您可以像这样解析和重组您的数据：
merged_result = [
"43f4cc160;Dan Ab;2021-06-17; 1 file changed, 10 insertions(+), 19 deletions(-)",
"6cbf2a8b3;Dan Ab;2021-06-15; 1 file changed, 14303 insertions(+)",
"c0a77029c;Dan Ab;2021-06-15; 1 file changed, 1 insertion(+), 1 deletion(-)",
"f283d7524;Dan Ab;2021-06-15; 1 file changed, 5260 deletions(-)",
"03c5314b4;Dan Ab;2021-06-15; 5 files changed, 5265 insertions(+), 12690 deletions(-)",
"daf38ecdf;Dan Ab;2020-12-11; 1 file changed, 8 insertions(+)",
"b5eabd543;Dan Ab;2020-10-14; 1 file changed, 17 insertions(+)",
"6d50a9d09;Dan Ab;2020-10-14; 43 files changed, 15740 insertions(+), 1 deletion(-)",
"6d50a9d09;Dan Ab;2020-10-14; Steak and Fries, no Salad",
]
节目：
import re
grouped = {}
pattern = r"(\d+) file[^,]*(?:\, (\d+) ins[^,]+)?(?:\, (\d+) del.+)?$"
for line in merged_result:
tag, name, date, changes = line.split(";", 3)
try:
# this will throw "NoneType" has no .groups() if not matched
files, inserts, deletes = re.search(pattern, changes).groups()
inserts, deletes = inserts or "0", deletes or "0"
except AttributeError as a:
print("Skipping: '",line,
"': cannot match data by regex to get changed/inserted/deleted\n", a)
continue
nameDict = grouped.setdefault(name, {})
dateDict = nameDict.setdefault(date, {})
dateDict.setdefault("commit_count", 0)
dateDict["commit_count"] += 1
commList = dateDict.setdefault("commit", [])
commList.append({"commit_hash": {"tag": tag, "files": files,
"lines_added": inserts, "lines_removed": deletes}})
print(grouped)
输出：
Skipping: ' 6d50a9d09;Dan Ab;2020-10-14; Steak and Fries, no Salad ': cannot match data by regex to get changed/inserted/deleted
{'Dan Ab': {'2021-06-17': {'commit_count': 1, 'commit': [{'commit_hash': {'tag': '43f4cc160', 'files': '1', 'lines_added': '10', 'lines_removed': '19'}}]}, '2021-06-15': {'commit_count': 4, 'commit': [{'commit_hash': {'tag': '6cbf2a8b3', 'files': '1', 'lines_added': '14303', 'lines_removed': '0'}}, {'commit_hash': {'tag': 'c0a77029c', 'files': '1', 'lines_added': '1', 'lines_removed': '1'}}, {'commit_hash': {'tag': 'f283d7524', 'files': '1', 'lines_added': '0', 'lines_removed': '5260'}}, {'commit_hash': {'tag': '03c5314b4', 'files': '5', 'lines_added': '5265', 'lines_removed': '12690'}}]}, '2020-12-11': {'commit_count': 1, 'commit': [{'commit_hash': {'tag': 'daf38ecdf', 'files': '1', 'lines_added': '8', 'lines_removed': '0'}}]}, '2020-10-14': {'commit_count': 2, 'commit': [{'commit_hash': {'tag': 'b5eabd543', 'files': '1', 'lines_added': '17', 'lines_removed': '0'}}, {'commit_hash': {'tag': '6d50a9d09', 'files': '43', 'lines_added': '15740', 'lines_removed': '1'}}]}}}
结果字典 reformatted（第一次点击 google）：
{
"Dan Ab":{
"2021-06-17":{
"commit_count":1,
"commit":[
{
"commit_hash":{
"tag":"43f4cc160",
"files":"1",
"lines_added":"10",
"lines_removed":"19"
}
}
]
},
"2021-06-15":{
"commit_count":4,
"commit":[
{
"commit_hash":{
"tag":"6cbf2a8b3",
"files":"1",
"lines_added":"14303",
"lines_removed":"0"
}
},
{
"commit_hash":{
"tag":"c0a77029c",
"files":"1",
"lines_added":"1",
"lines_removed":"1"
}
},
{
"commit_hash":{
"tag":"f283d7524",
"files":"1",
"lines_added":"0",
"lines_removed":"5260"
}
},
{
"commit_hash":{
"tag":"03c5314b4",
"files":"5",
"lines_added":"5265",
"lines_removed":"12690"
}
}
]
},
"2020-12-11":{
"commit_count":1,
"commit":[
{
"commit_hash":{
"tag":"daf38ecdf",
"files":"1",
"lines_added":"8",
"lines_removed":"0"
}
}
]
},
"2020-10-14":{
"commit_count":2,
"commit":[
{
"commit_hash":{
"tag":"b5eabd543",
"files":"1",
"lines_added":"17",
"lines_removed":"0"
}
},
{
"commit_hash":{
"tag":"6d50a9d09",
"files":"43",
"lines_added":"15740",
"lines_removed":"1"
}
}
]
}
}
}
r"(\d+) file[^,]*(?:\, (\d+) ins[^,]+)?(?:\, (\d+) del.+)?$" 的解释：
我正在使用 re.search 所以它不一定从字符串的开头开始 - 我正在寻找：
(\d+) file[^,] a number followed by "file" consuming
anything up to (excluding) the next ","
capturing the number in a group
(?:\, (\d+) ins[^,]+)?
(?:\, (\d+) del.+)? are similar: 0 to 1 occurence of ", "
followed by a captured number
followed by a space and some text
after "ins" we capture anyting up to excluding
the next ","
after "del" we simply capture anything
$ followed by end of string
如果不存在，可选组将导致 None，因此使用 inserts, deletes = inserts or "0", deletes or "0"
如果您需要更快的速度，您可以使用defaultdict()，但dict.setdefault 也可以。