Skip to content

feat: optimize the parsing code for speed #330

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 6 commits into from
Closed

feat: optimize the parsing code for speed #330

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
a996c96
optimize the parsing code for speed
APN-Pucky Oct 30, 2025
c99bf2d
splitlines failed
APN-Pucky Oct 30, 2025
8c9ac17
clean every 10th event
APN-Pucky Oct 30, 2025
fefc850
root clear idx counter should count idx manually
APN-Pucky Oct 30, 2025
d0fc700
update dict in one go
APN-Pucky Oct 30, 2025
753e01b
drop paranoid int(float()) conversion
APN-Pucky Oct 30, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
145 changes: 79 additions & 66 deletions src/pylhe/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -190,8 +190,8 @@ def fromstring(cls, string: str) -> "LHEEventInfo":
"""
values = string.split()
return cls(
nparticles=int(float(values[0])),
pid=int(float(values[1])),
nparticles=int(values[0]),
pid=int(values[1]),
weight=float(values[2]),
scale=float(values[3]),
aqed=float(values[4]),
Expand Down Expand Up @@ -256,12 +256,12 @@ def fromstring(cls, string: str) -> "LHEParticle":
"""
values = string.split()
return cls(
id=int(float(values[0])),
status=int(float(values[1])),
mother1=int(float(values[2])),
mother2=int(float(values[3])),
color1=int(float(values[4])),
color2=int(float(values[5])),
id=int(values[0]),
status=int(values[1]),
mother1=int(values[2]),
mother2=int(values[3]),
color1=int(values[4]),
color2=int(values[5]),
px=float(values[6]),
py=float(values[7]),
pz=float(values[8]),
Expand Down Expand Up @@ -354,16 +354,16 @@ def fromstring(cls, string: str) -> "LHEInitInfo":
"""
values = string.split()
return cls(
beamA=int(float(values[0])),
beamB=int(float(values[1])),
beamA=int(values[0]),
beamB=int(values[1]),
energyA=float(values[2]),
energyB=float(values[3]),
PDFgroupA=int(float(values[4])),
PDFgroupB=int(float(values[5])),
PDFsetA=int(float(values[6])),
PDFsetB=int(float(values[7])),
weightingStrategy=int(float(values[8])),
numProcesses=int(float(values[9])),
PDFgroupA=int(values[4]),
PDFgroupB=int(values[5]),
PDFsetA=int(values[6]),
PDFsetB=int(values[7]),
weightingStrategy=int(values[8]),
numProcesses=int(values[9]),
)


Expand Down Expand Up @@ -399,7 +399,7 @@ def fromstring(cls, string: str) -> "LHEProcInfo":
xSection=float(values[0]),
error=float(values[1]),
unitWeight=float(values[2]),
procId=int(float(values[3])),
procId=int(values[3]),
)


Expand Down Expand Up @@ -640,58 +640,71 @@ def _fromcontext(
with_attributes: bool,
) -> Iterator["LHEEvent"]:
index_map = _get_index_to_id_map(lheinit) if with_attributes else {}
for event, element in context:
if event == "end" and element.tag == "event":
if element.text is None:
err = "<event> block has no text."
raise ValueError(err)
idx = 0

data = element.text.strip().split("\n")
eventdata_str, particles_str = data[0], data[1:]

eventinfo = LHEEventInfo.fromstring(eventdata_str)
particles = [
LHEParticle.fromstring(p)
for p in particles_str
if not p.strip().startswith("#")
]

if with_attributes:
weights = {}
attrib = element.attrib
optional = [
p.strip() for p in particles_str if p.strip().startswith("#")
try:
for event, element in context:
if event == "end" and element.tag == "event":
idx = idx + 1
if element.text is None:
err = "<event> block has no text."
raise ValueError(err)

data = element.text.strip().split("\n")
eventdata_str = data[0]
particles_str = [p.strip() for p in data[1:]]

eventinfo = LHEEventInfo.fromstring(eventdata_str)
particles = [
LHEParticle.fromstring(p)
for p in particles_str
if not p.startswith("#")
]

for sub in element:
if sub.tag == "weights":
if sub.text is None:
err = "<weights> block has no text."
raise ValueError(err)
for i, w in enumerate(sub.text.split()):
if w and index_map[i] not in weights:
weights[index_map[i]] = float(w)
elif sub.tag == "rwgt":
for r in sub:
if r.tag == "wgt":
if r.text is None:
err = "<wgt> block has no text."
raise ValueError(err)
weights[r.attrib["id"]] = float(r.text.strip())

yield LHEEvent(
eventinfo=eventinfo,
particles=particles,
weights=weights,
attributes=attrib,
optional=optional,
)
else:
yield LHEEvent(eventinfo, particles)

# Clear memory
element.clear()
root.clear()
if with_attributes:
weights = {}
attrib = element.attrib
optional = [p for p in particles_str if p.startswith("#")]

for sub in element:
if sub.tag == "weights":
if sub.text is None:
err = "<weights> block has no text."
raise ValueError(err)
weights.update(
{
index_map[i]: float(w)
for i, w in enumerate(
sub.text.split()[::-1]
) # reversed order to only keep the first occurrence in case of duplicates
if w and index_map[i] not in weights
}
)
elif sub.tag == "rwgt":
for r in sub:
if r.tag == "wgt":
if r.text is None:
err = "<wgt> block has no text."
raise ValueError(err)
weights[r.attrib["id"]] = float(r.text.strip())

yield LHEEvent(
eventinfo=eventinfo,
particles=particles,
weights=weights,
attributes=attrib,
optional=optional,
)
else:
yield LHEEvent(eventinfo, particles)

# Clear memory
element.clear()
# Clear the root every 100 elements
if idx % 100 == 0:
root.clear()
finally:
root.clear()

@property
def graph(self) -> graphviz.Digraph:
Expand Down