Cathode on Pin 1, Cap + on Pin 1
Not sure who needs to read this, but that is the IPC convention — IPC-7351B for surface mount, IPC-7251 for through-hole. Footprints are built that way, so naturally the symbols have to be done the same way. A diode cathode goes on pin 1, and on a polarized capacitor the + goes on pin 1.
It may seem like it does not make a difference. In manufacturing, debug and test it does.
Why pin 1
Diode polarity and orientation are read from the schematic. Zero Component Orientation (ZCO) in IPC-7351B and IPC-7251 requires pin 1, the cathode, to be on the left with the footprint at zero rotation. Everything after that is built on the assumption: the machine that places the part, the equipment that tests it, and the engineer who later reads a netlist diagnostic.
What it looks like in a report
The diode section is often where this surfaces for the first time. Two reset diodes in a real design, both flagged High:
Symbol modeling error: the symbol places the cathode on pin 2, but IPC-7351B (SMT) / IPC-7251 (through-hole) require pin 1 as the cathode. Pick-and-place machines and ATE are oriented to pin 1 at zero body rotation. Customizing the footprint to overcome this error still leaves potential problems, as netlist diagnostics refer to pin 1, which is familiar as the cathode, and this design does it differently. Findings on D15 and D16, LRB520S-30T1G, in the RESET section of the Raspberry Pi CM4S carrier by Petr Dvorak.
A symbol that does not meet ZCO gets flagged because of what can follow from it: the layout may carry an error, assembly may fit the part backwards, or someone less experienced may wrongly indict a part that was correct all along. Staying consistent is the least-expensive path for everyone who handles the PCB.
See a Raspberry Pi Compute Module carrier board reviewed end to end — CM4, CM4S and CM5